@article{liu_berman_dodson_park_zahabi_huang_ruiz_kaber_2024, title={Human-Centered Evaluation of EMG-Based Upper-Limb Prosthetic Control Modes}, volume={4}, ISSN={["2168-2305"]}, url={http://dx.doi.org/10.1109/thms.2024.3381094}, DOI={10.1109/thms.2024.3381094}, abstractNote={The aim of this study was to experimentally test the effects of different electromyographic-based prosthetic control modes on user task performance, cognitive workload, and perceived usability to inform further human-centered design and application of these prosthetic control interfaces. We recruited 30 able-bodied participants for a between-subjects comparison of three control modes: direct control (DC), pattern recognition (PR), and continuous control (CC). Multiple human-centered evaluations were used, including task performance, cognitive workload, and usability assessments. To ensure that the results were not task-dependent, this study used two different test tasks, including the clothespin relocation task and Southampton hand assessment procedure-door handle task. Results revealed performance with each control mode to vary among tasks. When the task had high-angle adjustment accuracy requirements, the PR control outperformed DC. For cognitive workload, the CC mode was superior to DC in reducing user load across tasks. Both CC and PR control appear to be effective alternatives to DC in terms of task performance and cognitive load. Furthermore, we observed that, when comparing control modes, multitask testing and multifaceted evaluations are critical to avoid task-induced or method-induced evaluation bias. Hence, future studies with larger samples and different designs will be needed to expand the understanding of prosthetic device features and workload relationships.}, journal={IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS}, author={Liu, Yunmei and Berman, Joseph and Dodson, Albert and Park, Junho and Zahabi, Maryam and Huang, He and Ruiz, Jaime and Kaber, David B.}, year={2024}, month={Apr} }
 @misc{huang_hargrove_ortiz-catalan_sensinger_2024, title={Integrating Upper-Limb Prostheses with the Human Body: Technology Advances, Readiness, and Roles in Human-Prosthesis Interaction}, volume={26}, ISSN={["1545-4274"]}, url={http://dx.doi.org/10.1146/annurev-bioeng-110222-095816}, DOI={10.1146/annurev-bioeng-110222-095816}, abstractNote={Significant advances in bionic prosthetics have occurred in the past two decades. The field's rapid expansion has yielded many exciting technologies that can enhance the physical, functional, and cognitive integration of a prosthetic limb with a human. We review advances in the engineering of prosthetic devices and their interfaces with the human nervous system, as well as various surgical techniques for altering human neuromusculoskeletal systems for seamless human-prosthesis integration. We discuss significant advancements in research and clinical translation, focusing on upper limbprosthetics since they heavily rely on user intent for daily operation, although many discussed technologies have been extended to lower limb prostheses as well. In addition, our review emphasizes the roles of advanced prosthetics technologies in complex interactions with humans and the technology readiness levels (TRLs) of individual research advances. Finally, we discuss current gaps and controversies in the field and point out future research directions, guided by TRLs.}, journal={ANNUAL REVIEW OF BIOMEDICAL ENGINEERING}, author={Huang, He and Hargrove, Levi J. and Ortiz-Catalan, Max and Sensinger, Jonathon W.}, year={2024}, pages={503–528} }
 @article{rubin_hinson_saul_filer_hu_huang_2024, title={Modified motor unit properties in residual muscle following transtibial amputation}, volume={21}, ISSN={["1741-2552"]}, DOI={10.1088/1741-2552/ad1ac2}, abstractNote={Abstract
               
                  Objective. Neural signals in residual muscles of amputated limbs are frequently decoded to control powered prostheses. Yet myoelectric controllers assume muscle activities of residual muscles are similar to that of intact muscles. This study sought to understand potential changes to motor unit (MU) properties after limb amputation. Approach. Six people with unilateral transtibial amputation were recruited. Surface electromyogram (EMG) of residual and intact tibialis anterior (TA) and gastrocnemius (GA) muscles were recorded while subjects traced profiles targeting up to 20% and 35% of maximum activation for each muscle (isometric for intact limbs). EMG was decomposed into groups of MU spike trains. MU recruitment thresholds, action potential amplitudes (MU size), and firing rates were correlated to model Henneman’s size principle, the onion-skin phenomenon, and rate-size associations. Organization (correlation) and modulation (rates of change) of relations were compared between intact and residual muscles. Main results. The residual TA exhibited significantly lower correlation and flatter slopes in the size principle and onion-skin, and each outcome covaried between the MU relations. The residual GA was unaffected for most subjects. Subjects trained prior with myoelectric prostheses had minimally affected slopes in the TA. Rate-size association correlations were preserved, but both residual muscles exhibited flatter decay rates. Significance. We showed peripheral neuromuscular damage also leads to spinal-level functional reorganizations. Our findings suggest models of MU recruitment and discharge patterns for residual muscle EMG generation need reparameterization to account for disturbances observed. In the future, tracking MU pool adaptations may also provide a biomarker of neuromuscular control to aid training with myoelectric prostheses.}, number={1}, journal={JOURNAL OF NEURAL ENGINEERING}, author={Rubin, Noah and Hinson, Robert and Saul, Katherine and Filer, William and Hu, Xiaogang and Huang, He}, year={2024}, month={Feb} }
 @misc{valero-cuevas_finley_orsborn_fung_hicks_huang_reinkensmeyer_schweighofer_weber_steele_2024, title={NSF DARE-Transforming modeling in neurorehabilitation: Four threads for catalyzing progress}, volume={21}, ISSN={["1743-0003"]}, url={http://dx.doi.org/10.1186/s12984-024-01324-x}, DOI={10.1186/s12984-024-01324-x}, abstractNote={We present an overview of the Conference on Transformative Opportunities for Modeling in Neurorehabilitation held in March 2023. It was supported by the Disability and Rehabilitation Engineering (DARE) program from the National Science Foundation's Engineering Biology and Health Cluster. The conference brought together experts and trainees from around the world to discuss critical questions, challenges, and opportunities at the intersection of computational modeling and neurorehabilitation to understand, optimize, and improve clinical translation of neurorehabilitation. We organized the conference around four key, relevant, and promising Focus Areas for modeling: Adaptation & Plasticity, Personalization, Human-Device Interactions, and Modeling 'In-the-Wild'. We identified four common threads across the Focus Areas that, if addressed, can catalyze progress in the short, medium, and long terms. These were: (i) the need to capture and curate appropriate and useful data necessary to develop, validate, and deploy useful computational models (ii) the need to create multi-scale models that span the personalization spectrum from individuals to populations, and from cellular to behavioral levels (iii) the need for algorithms that extract as much information from available data, while requiring as little data as possible from each client (iv) the insistence on leveraging readily available sensors and data systems to push model-driven treatments from the lab, and into the clinic, home, workplace, and community. The conference archive can be found at (dare2023.usc.edu). These topics are also extended by three perspective papers prepared by trainees and junior faculty, clinician researchers, and federal funding agency representatives who attended the conference.}, number={1}, journal={JOURNAL OF NEUROENGINEERING AND REHABILITATION}, author={Valero-Cuevas, Francisco J. and Finley, James and Orsborn, Amy and Fung, Natalie and Hicks, Jennifer L. and Huang, He and Reinkensmeyer, David and Schweighofer, Nicolas and Weber, Douglas and Steele, Katherine M.}, year={2024}, month={Apr} }
 @article{gao_si_huang_2023, title={Reinforcement Learning Control With Knowledge Shaping}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85149366397&partnerID=MN8TOARS}, DOI={10.1109/TNNLS.2023.3243631}, abstractNote={We aim at creating a transfer reinforcement learning framework that allows the design of learning controllers to leverage prior knowledge extracted from previously learned tasks and previous data to improve the learning performance of new tasks. Toward this goal, we formalize knowledge transfer by expressing knowledge in the value function in our problem construct, which is referred to as reinforcement learning with knowledge shaping (RL-KS). Unlike most transfer learning studies that are empirical in nature, our results include not only simulation verifications but also an analysis of algorithm convergence and solution optimality. Also different from the well-established potential-based reward shaping methods which are built on proofs of policy invariance, our RL-KS approach allows us to advance toward a new theoretical result on positive knowledge transfer. Furthermore, our contributions include two principled ways that cover a range of realization schemes to represent prior knowledge in RL-KS. We provide extensive and systematic evaluations of the proposed RL-KS method. The evaluation environments not only include classical RL benchmark problems but also include a challenging task of real-time control of a robotic lower limb with a human user in the loop.}, journal={IEEE Transactions on Neural Networks and Learning Systems}, author={Gao, X. and Si, J. and Huang, H.}, year={2023} }
 @article{zhang_si_tu_li_lewek_huang_2024, title={Toward Task-Independent Optimal Adaptive Control of a Hip Exoskeleton for Locomotion Assistance in Neurorehabilitation}, volume={9}, ISSN={["2168-2232"]}, DOI={10.1109/TSMC.2024.3454556}, journal={IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS}, author={Zhang, Qiang and Si, Jennie and Tu, Xikai and Li, Minhan and Lewek, Michael D. and Huang, He}, year={2024}, month={Sep} }
 @article{driscoll_liu_huang_2023, title={1-D Manual Tracing Based on a High Density Haptic Stimulation Grid - a Pilot Effort}, ISSN={["2835-9518"]}, url={http://dx.doi.org/10.1109/whc56415.2023.10224505}, DOI={10.1109/whc56415.2023.10224505}, abstractNote={Lower limb amputees lack the neurological path-ways needed for perception of how their prosthetic limbs are interacting with the environment, leading to a lack of confidence in their devices and reduced balancing capabilities. Sensory substitution methods, such as vibrotactile and electrotactile feedback applied to unaffected body segments offer a potential way to restore some of the lost information pathways. While high resolution haptic stimulation grids have become commercially available, few studies have tried to make use of these devices to provide more intuitive sensory substitution methods. This study developed an encoding approach, which is based on the illusory “phantom actuator” phenomenon, to convert 1-D position information to a wearer through a bHaptics Tactsuit. By evaluating performance of 1-D manual tracking task among 14 participants under the proposed approach and a traditional amplitude modulation approach, we demonstrated an improvement of velocity tracing accuracy (p=0.0375) with the proposed approach, although the proposed approach did not lead to significant improvement in the position tracing accuracy.}, journal={2023 IEEE WORLD HAPTICS CONFERENCE, WHC}, publisher={IEEE}, author={Driscoll, Brendan and Liu, Ming and Huang, He}, year={2023}, pages={375–381} }
 @article{alili_nalam_li_liu_feng_si_huang_2023, title={A Novel Framework to Facilitate User Preferred Tuning for a Robotic Knee Prosthesis}, volume={31}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85147231065&partnerID=MN8TOARS}, DOI={10.1109/TNSRE.2023.3236217}, abstractNote={The tuning of robotic prosthesis control is essential to provide personalized assistance to individual prosthesis users. Emerging automatic tuning algorithms have shown promise to ease the device personalization procedure. However, very few automatic tuning algorithms consider the user preference as the tuning goal, which may limit the adoptability of the robotic prosthesis. In this study, we propose and evaluate a novel prosthesis control tuning framework for a robotic knee prosthesis, which could enable user preferred robot behavior in the device tuning process. The framework consists of 1) a User-Controlled Interface that allows the user to select their preferred knee kinematics in gait and 2) a reinforcement learning-based algorithm for tuning high-dimension prosthesis control parameters to meet the desired knee kinematics. We evaluated the performance of the framework along with usability of the developed user interface. In addition, we used the developed framework to investigate whether amputee users can exhibit a preference between different profiles during walking and whether they can differentiate between their preferred profile and other profiles when blinded. The results showed effectiveness of our developed framework in tuning 12 robotic knee prosthesis control parameters while meeting the user-selected knee kinematics. A blinded comparative study showed that users can accurately and consistently identify their preferred prosthetic control knee profile. Further, we preliminarily examined gait biomechanics of the prosthesis users when walking with different prosthesis control and did not find clear difference between walking with preferred prosthesis control and when walking with normative gait control parameters. This study may inform future translation of this novel prosthesis tuning framework for home or clinical use.}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Alili, Abbas and Nalam, Varun and Li, Minhan and Liu, Ming and Feng, Jing and Si, Jennie and Huang, He}, year={2023}, pages={895–903} }
 @article{zhang_tu_si_lewek_huang_2023, title={A Robotic Assistance Personalization Control Approach of Hip Exoskeletons for Gait Symmetry Improvement}, ISSN={["2153-0858"]}, DOI={10.1109/IROS55552.2023.10341440}, abstractNote={Healthy human locomotion functions with good gait symmetry depend on rhythmic coordination of the left and right legs, which can be deteriorated by neurological disorders like stroke and spinal cord injury. Powered exoskeletons are promising devices to improve impaired people's locomotion functions, like gait symmetry. However, given higher uncertainties and the time-varying nature of human-robot interaction, providing personalized robotic assistance from exoskeletons to achieve the best gait symmetry is challenging, especially for people with neurological disorders. In this paper, we propose a hierarchical control framework for a bilateral hip exoskeleton to provide the adaptive optimal hip joint assistance with a control objective of imposing the desired gait symmetry during walking. Three control levels are included in the hierarchical framework, including the high-level control to tune three control parameters based on a policy iteration reinforcement learning approach, the middle-level control to define the desired assistive torque profile based on a delayed output feedback control method, and the low-level control to achieve a good torque trajectory tracking performance. To evaluate the feasibility of the proposed control framework, five healthy young participants are recruited for treadmill walking experiments, where an artificial gait asymmetry is imitated as the hemiparesis post-stroke, and only the ‘paretic’ hip joint is controlled with the proposed framework. The pilot experimental studies demonstrate that the hierarchical control framework for the hip exoskeleton successfully (asymmetry index from 8.8% to − 0.5%) and efficiently (less than 4 minutes) achieved the desired gait symmetry by providing adaptive optimal assistance on the ‘paretic’ hip joint.}, journal={2023 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS)}, author={Zhang, Qiang and Tu, Xikai and Si, Jennie and Lewek, Michael D. and Huang, He}, year={2023}, pages={6125–6132} }
 @article{yu_nalam_alili_huang_2023, title={A Wearable Robotic Rehabilitation System for Neuro-rehabilitation Aimed at Enhancing Mediolateral Balance}, ISSN={["2153-0858"]}, DOI={10.1109/IROS55552.2023.10341735}, abstractNote={There is increasing evidence of the role of compromised mediolateral balance in falls and the need for rehabilitation specifically focused on mediolateral direction for various populations with motor deficits. To address this need, we have developed a neurorehabilitation platform by integrating a wearable robotic hip abduction-adduction exoskeleton with a visual interface. The platform is expected to influence and rehabilitate the underlying visuomotor mechanisms in individuals by having users perform motion tasks based on visual feedback while the robot applies various controlled resistances governed by the admittance controller implemented in the robot. A preliminary study was performed on 3 non disabled individuals to analyze the performance of the system and observe any adaptation in hip joint kinematics and kinetics as a result of the visuomotor training under 4 different admittance conditions. All three subjects exhibited increased consistency of motion during training and interlimb coordination to achieve motion tasks, demonstrating the utility of the system. Further analysis of observed human-robot torque interactions and electromyography (EMG) signals, and its implication in neurorehabilitation aimed at populations suffering from chronic stroke are discussed.}, journal={2023 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, IROS}, author={Yu, Zhenyuan and Nalam, Varun and Alili, Abbas and Huang, He}, year={2023}, pages={155–160} }
 @article{alili_fleming_nalam_liu_dean_huang_2024, title={Abduction/Adduction Assistance From Powered Hip Exoskeleton Enables Modulation of User Step Width During Walking}, volume={71}, ISSN={["1558-2531"]}, url={http://dx.doi.org/10.1109/tbme.2023.3301444}, DOI={10.1109/TBME.2023.3301444}, abstractNote={Using wearable robotics to modulate step width in normal walking for enhanced mediolateral balance has not been demonstrated in the field. We designed a bilateral hip exoskeleton with admittance control to power hip abduction and adduction to modulate step width. Objective: As the first step to show its potential, the objective of this study was to investigate how human's step width reacted to hip exoskeleton's admittance control parameter changes during walking. Methods: Ten non-disabled individuals walked on a treadmill at a self-selected speed, while wearing our bilateral robotic hip exoskeleton. We used two equilibrium positions to define the direction of assistance. We studied the influence of multiple stiffness values in the admittance control on the participants’ step width, step length, and electromyographic (EMG) activity of the gluteus medius. Results: Step width were significantly modulated by the change of stiffness in exoskeleton control, while step length did not show significant changes. When the stiffness changed from zero to our studied stiffness values, the participants’ step width started to modulate immediately. Within 4 consecutive heel strikes right after a stiffness change, the step width showed a significant change. Interestingly, EMG activity of the gluteus medius did not change significantly regardless the applied stiffness and powered direction. Conclusion: Tuning of stiffness in admittance control of a hip exoskeleton, acting in mediolateral direction, can be a viable way for controlling step width in normal walking. Unvaried gluteus medius activity indicates that the increase in step width were mainly caused by the assistive torque applied by the exoskeleton. Significance: Our study results pave a new way for future design and control of wearable robotics in enhancing mediolateral walking balance for various rehabilitation applications.}, number={1}, journal={IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Alili, Abbas and Fleming, Aaron and Nalam, Varun and Liu, Ming and Dean, Jesse and Huang, He}, year={2024}, month={Jan}, pages={334–342} }
 @article{hong_zhao_berman_chi_li_huang_yin_2023, title={Angle-programmed tendril-like trajectories enable a multifunctional gripper with ultradelicacy, ultrastrength, and ultraprecision}, volume={14}, ISSN={["2041-1723"]}, DOI={10.1038/s41467-023-39741-6}, abstractNote={AbstractAchieving multicapability in a single soft gripper for handling ultrasoft, ultrathin, and ultraheavy objects is challenging due to the tradeoff between compliance, strength, and precision. Here, combining experiments, theory, and simulation, we report utilizing angle-programmed tendril-like grasping trajectories for an ultragentle yet ultrastrong and ultraprecise gripper. The single gripper can delicately grasp fragile liquids with minimal contact pressure (0.05 kPa), lift objects 16,000 times its own weight, and precisely grasp ultrathin, flexible objects like 4-μm-thick sheets and 2-μm-diameter microfibers on flat surfaces, all with a high success rate. Its scalable and material-independent design allows for biodegradable noninvasive grippers made from natural leaves. Explicitly controlled trajectories facilitate its integration with robotic arms and prostheses for challenging tasks, including picking grapes, opening zippers, folding clothes, and turning pages. This work showcases soft grippers excelling in extreme scenarios with potential applications in agriculture, food processing, prosthesis, biomedicine, minimally invasive surgeries, and deep-sea exploration.}, number={1}, journal={NATURE COMMUNICATIONS}, author={Hong, Yaoye and Zhao, Yao and Berman, Joseph and Chi, Yinding and Li, Yanbin and Huang, He and Yin, Jie}, year={2023}, month={Aug} }
 @article{rubin_hinson_saul_hu_huang_2023, title={Ankle Torque Estimation With Motor Unit Discharges in Residual Muscles Following Lower-Limb Amputation}, volume={31}, ISSN={["1558-0210"]}, url={http://dx.doi.org/10.1109/tnsre.2023.3336543}, DOI={10.1109/TNSRE.2023.3336543}, abstractNote={There has been increased interest in using residual muscle activity for neural control of powered lower-limb prostheses. However, only surface electromyography (EMG)-based decoders have been investigated. This study aims to investigate the potential of using motor unit (MU)-based decoding methods as an alternative to EMG-based intent recognition for ankle torque estimation. Eight people without amputation (NON) and seven people with amputation (AMP) participated in the experiments. Subjects conducted isometric dorsi- and plantarflexion with their intact limb by tracing desired muscle activity of the tibialis anterior (TA) and gastrocnemius (GA) while ankle torque was recorded. To match phantom limb and intact limb activity, AMP mirrored muscle activation with their residual TA and GA. We compared neuromuscular decoders (linear regression) for ankle joint torque estimation based on 1) EMG amplitude (aEMG), 2) MU firing frequencies representing neural drive (ND), and 3) MU firings convolved with modeled twitch forces (MUDrive). In addition, sensitivity analysis and dimensionality reduction of optimization were performed on the MUDrive method to further improve its practical value. Our results suggest MUDrive significantly outperforms (lower root-mean-square error) EMG and ND methods in muscles of NON, as well as both intact and residual muscles of AMP. Reducing the number of optimized MUDrive parameters degraded performance. Even so, optimization computational time was reduced and MUDrive still outperformed aEMG. Our outcomes indicate integrating MU discharges with modeled biomechanical outputs may provide a more accurate torque control signal than direct EMG control of assistive, lower-limb devices, such as exoskeletons and powered prostheses.}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Rubin, Noah and Hinson, Robert and Saul, Katherine and Hu, Xiaogang and Huang, He}, year={2023}, pages={4821–4830} }
 @article{yip_salcudean_goldberg_althoefer_menciassi_opfermann_krieger_swaminathan_walsh_huang_et al._2023, title={Artificial intelligence meets medical robotics}, volume={381}, ISSN={["1095-9203"]}, DOI={10.1126/science.adj3312}, abstractNote={
            Artificial intelligence (AI) applications in medical robots are bringing a new era to medicine. Advanced medical robots can perform diagnostic and surgical procedures, aid rehabilitation, and provide symbiotic prosthetics to replace limbs. The technology used in these devices, including computer vision, medical image analysis, haptics, navigation, precise manipulation, and machine learning (ML) , could allow autonomous robots to carry out diagnostic imaging, remote surgery, surgical subtasks, or even entire surgical procedures. Moreover, AI in rehabilitation devices and advanced prosthetics can provide individualized support, as well as improved functionality and mobility (see the figure). The combination of extraordinary advances in robotics, medicine, materials science, and computing could bring safer, more efficient, and more widely available patient care in the future.
            –Gemma K. Alderton
          }, number={6654}, journal={SCIENCE}, author={Yip, Michael and Salcudean, Septimiu and Goldberg, Ken and Althoefer, Kaspar and Menciassi, Arianna and Opfermann, Justin D. D. and Krieger, Axel and Swaminathan, Krithika and Walsh, Conor J. J. and Huang, He and et al.}, year={2023}, month={Jul}, pages={141–146} }
 @article{park_berman_dodson_liu_armstrong_huang_kaber_ruiz_zahabi_2023, title={Assessing workload in using electromyography (EMG)-based prostheses}, volume={6}, ISSN={["1366-5847"]}, DOI={10.1080/00140139.2023.2221413}, abstractNote={Using prosthetic devices requires substantial cognitive workload. This study investigated classification models for assessing cognitive workload in electromyography (EMG)-based prosthetic devices with various types of input features including eye-tracking measures, task performance, and cognitive performance model (CPM) outcomes. Features selection algorithm, hyperparameter tuning with grid search, and k-fold cross validation were applied to select the most important features and find the optimal models. Classification accuracy, area under the receiver operation characteristic curve (AUC), precision, recall, and F1 scores were calculated to compare models' performance. The findings suggested that task performance measures, pupillometry data, and CPM outcomes, combined with the naïve bayes (NB) and random forest (RF) algorithms, are most promising for classifying cognitive workload. The proposed algorithms can help manufacturers/clinicians predict cognitive workload of future EMG-based prosthetic devices in early design phases.}, journal={ERGONOMICS}, author={Park, Junho and Berman, Joseph and Dodson, Albert and Liu, Yunmei and Armstrong, Matthew and Huang, He and Kaber, David and Ruiz, Jaime and Zahabi, Maryam}, year={2023}, month={Jun} }
 @article{alili_nalam_fleming_liu_dean_huang_2023, title={Closed-Loop Feedback Control of Human Step Width During Walking by Mediolaterally Acting Robotic Hip Exoskeleton}, ISSN={["2153-0858"]}, DOI={10.1109/IROS55552.2023.10342127}, abstractNote={Maintaining balance during gait in the mediolateral direction requires more active motor control than in the anteroposterior direction. Step width modulation is a key strategy used by healthy individuals to achieve mediolateral walking balance, but it can be disrupted in populations with poor sensorimotor integration and weak hip abductors, such as the elderly, stroke patients, and people with lower limb amputation. Wearable hip exoskeletons have the potential to serve as assistive or rehabilitation devices for these populations, but there has been limited research on their appropriate usage. In this study, we successfully demonstrated the feasibility of controlling step width using a mediolaterally acting robotic hip exoskeleton. We were able to effectively adjust the user's step width by increasing or decreasing it to predefined targets through the regulation of admittance control parameters governing the device. The maximum average error to increase or decrease the step width was 1.2 cm. This research has the potential to facilitate the development of assistive and rehabilitation applications focused on enhancing the mediolateral gait balance of individuals with neurological impairments, elderly individuals, and amputees via the control of step width.}, journal={2023 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS)}, author={Alili, Abbas and Nalam, Varun and Fleming, Aaron and Liu, Ming and Dean, Jesse and Huang, Helen}, year={2023}, pages={6097–6102} }
 @inproceedings{park_music_delgado_berman_dodson_liu_ruiz_huang_kaber_zahabi_2023, title={Cognitive Workload and Usability of Virtual Reality Simulation for Prosthesis Training}, url={http://dx.doi.org/10.1109/smc53992.2023.10394286}, DOI={10.1109/smc53992.2023.10394286}, abstractNote={Amputees use prosthetic devices to perform activities of daily living. However, some users reject their devices due to the lack of usability or high cognitive workload. Although virtual reality has been studied in this domain for training purposes, there has not been any investigation on usability and cognitive workload of using virtual reality simulations for training of prosthetic devices. The objective of this study was to compare cognitive workload and usability of using virtual reality-based simulation of electromyography based prosthetic devices and physical devices. The findings suggested that using virtual reality simulations were helpful in reducing cognitive workload and increasing perceived usability of prosthetic devices.}, booktitle={2023 IEEE International Conference on Systems, Man, and Cybernetics (SMC)}, author={Park, Junho and Music, Austin and Delgado, Daniel and Berman, Joseph and Dodson, Albert and Liu, Yunmei and Ruiz, Jaime and Huang, He and Kaber, David and Zahabi, Maryam}, year={2023}, month={Oct} }
 @article{naseri_liu_lee_huang_2023, title={Development and Online Validation of an Intrinsic Fault Detector for a Powered Robotic Knee Prosthesis}, ISSN={["2153-0858"]}, DOI={10.1109/IROS55552.2023.10342433}, abstractNote={Robotic prosthetic legs have the potential to significantly improve the quality of life for lower limb amputees to perform locomotion in various environments and task conditions. However, these devices lack the capability to recover from internal intrinsic control faults, which can lead to harmful consequences affecting the user's gait performance and eroding trust in these robotic devices. Therefore, a reliable fault detection system is necessary to detect intrinsic faults in a timely manner and provide a compensatory response to mitigate their effects. This paper focuses on designing an active fault detector for a robotic knee prosthesis and demonstrates its effectiveness in real time. The developed system utilizes a Gaussian Process model to estimate knee angular velocity, which is sensitive to intrinsic faults and relies on the difference between estimated velocity and the actual measurement to detect internal control faults. In an offline analysis, the developed detector demonstrated a higher detection rate, lower false alarm ratio, and faster detection time compared with the two approaches reported previously. An online demonstration was also conducted with a unilateral amputee participant and showed performance similar to that of offline analysis. We expect that this detector can be integrated into a fault tolerance strategy to enhance the reliability and safety of robotic prosthetic legs, enabling users to perform their everyday tasks with greater confidence.}, journal={2023 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS, IROS}, author={Naseri, Amirreza and Liu, Ming and Lee, I-Chieh and Huang, Helen}, year={2023}, pages={2158–2164} }
 @inproceedings{yuan_bai_alili_liu_feng_huang_2023, title={Finding a Natural Fit: A Thematic Analysis of Amputees’ Prosthesis Setting Preferences during User-Guided Auto-Tuning}, url={http://dx.doi.org/10.1177/21695067231216121}, DOI={10.1177/21695067231216121}, abstractNote={Amputees’ preferences for prosthesis settings are critical not only for their psychological well-being but also for long-term adherence to device adoption and health. Although active lower-limb prostheses can provide enhanced functionality than passive devices, little is known about the mechanism of preferences for settings in active devices. Therefore, a think-aloud study was conducted on three amputees to unravel their preferences for a powered robotic knee prosthesis during user-guided auto-tuning. The inductive thematic analysis revealed that amputee patients were more likely to use their own passive device rather than the intact leg as the reference for the natural walking that they were looking for in the powered device. There were large individual differences in factors influencing naturalness. The mental optimization of preference decisions was mostly based on the noticeableness of the differences between knee profiles. The implications on future design and research in active prostheses were discussed.}, booktitle={Proceedings of the Human Factors and Ergonomics Society Annual Meeting}, author={Yuan, Jing and Bai, Xiaolu and Alili, Abbas and Liu, Ming and Feng, Jing and Huang, He}, year={2023}, month={Sep} }
 @article{berman_hinson_lee_huang_2023, title={Harnessing Machine Learning and Physiological Knowledge for a Novel EMG-Based Neural-Machine Interface}, volume={70}, ISSN={["1558-2531"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85139470198&partnerID=MN8TOARS}, DOI={10.1109/TBME.2022.3210892}, abstractNote={Objective: In this study, we aimed to develop a novel electromyography (EMG)-based neural machine interface (NMI), called the Neural Network-Musculoskeletal hybrid Model (N2M2), to decode continuous joint angles. Our approach combines the concepts of machine learning and musculoskeletal modeling. Methods: We compared our novel design with a musculoskeletal model (MM) and 2 continuous EMG decoders based on artificial neural networks (ANNs): multilayer perceptrons (MLPs) and nonlinear autoregressive neural networks with exogenous inputs (NARX networks). EMG and joint kinematics data were collected from 10 non-disabled and 1 transradial amputee subject. The offline performance tested across 3 different conditions (i.e., varied arm postures, shifted electrode locations, and noise-contaminated EMG signals) and online performance for a virtual postural matching task was quantified. Finally, we implemented the N2M2 to operate a prosthetic hand and tested functional task performance. Results: The N2M2 made more accurate predictions than the MLP in all postures and electrode locations (p < 0.003). For estimated MCP joint angles, the N2M2 was less sensitive to noisy EMG signals than the MM or NARX network with respect to error (p < 0.032) as well as the NARX network with respect to correlation (p = 0.007). Additionally, the N2M2 had better online task performance than the NARX network (p ≤ 0.030). Conclusion: Overall, we have found that combining the concepts of machine learning and musculoskeletal modeling has resulted in a more robust joint kinematics decoder than either concept individually. Significance: The outcome of this study may result in a novel, highly reliable controller for powered prosthetic hands.}, number={4}, journal={IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING}, author={Berman, Joseph and Hinson, Robert and Lee, I-Chieh and Huang, He}, year={2023}, month={Apr}, pages={1125–1136} }
 @article{li_liu_si_stallrich_huang_2023, title={Hierarchical Optimization for Control of Robotic Knee Prostheses Toward Improved Symmetry of Propulsive Impulse}, volume={70}, ISSN={["1558-2531"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85142785855&partnerID=MN8TOARS}, DOI={10.1109/TBME.2022.3224026}, abstractNote={Automatically personalizing complex control of robotic prostheses to improve gait performance, such as gait symmetry, is challenging. Recently, human-in-the-loop (HIL) optimization and reinforcement learning (RL) have shown promise in achieving optimized control of wearable robots for each individual user. However, HIL optimization methods lack scalability for high-dimensional space, while RL has mostly focused on optimizing robot kinematic performance. Thus, we propose a novel hierarchical framework to personalize robotic knee prosthesis control and improve overall gait performance. Specifically, in this study the framework was implemented to simultaneously design target knee kinematics and tune 12 impedance control parameters for improved symmetry of propulsive impulse in walking. In our proposed framework, HIL optimization is used to identify an optimal target knee kinematics with respect to symmetry improvement, while RL is leveraged to yield an optimal policy for tuning impedance parameters in high-dimensional space to match the kinematics target. The proposed framework was validated on human subjects, walking with robotic knee prosthesis. The results showed that our design successfully shaped the target knee kinematics as well as configured 12 impedance control parameters to improve propulsive impulse symmetry of the human users. The knee kinematics that yielded best propulsion symmetry did not preserve the normative knee kinematics profile observed in non-disabled individuals, suggesting that restoration of normative joint biomechanics in walking does not necessarily optimize the gait performance of human-prosthesis systems. This new framework for prosthesis control personalization may be extended to other wearable devices or different gait performance optimization goals in the future.}, number={5}, journal={IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING}, author={Li, Minhan and Liu, Wentao and Si, Jennie and Stallrich, Jonathan W. and Huang, He}, year={2023}, month={May}, pages={1634–1642} }
 @article{naseri_lee_huang_liu_2023, title={Investigating the Association of Quantitative Gait Stability Metrics With User Perception of Gait Interruption Due to Control Faults During Human-Prosthesis Interaction}, volume={31}, ISSN={["1558-0210"]}, DOI={10.1109/TNSRE.2023.3328877}, abstractNote={This study aims to compare the association of different gait stability metrics with the prosthesis users’ perception of their own gait stability. Lack of perceived confidence on the device functionality can influence the gait pattern, level of daily activities, and overall quality of life for individuals with lower limb motor deficits. However, the perception of gait stability is subjective and difficult to acquire online. The quantitative gait stability metrics can be objectively measured and monitored using wearable sensors; however, objective measurements of gait stability associated with human’s perception of their own gait stability has rarely been reported. By identifying quantitative measurements that associate with users’ perceptions, we can gain a more accurate and comprehensive understanding of an individual’s perceived functional outcomes of assistive devices such as prostheses. To achieve our research goal, experiments were conducted to artificially apply internal disturbances in the powered prosthesis while the prosthetic users performed level ground walking. We monitored and compared multiple gait stability metrics and a local measurement to the users’ reported perception of their own gait stability. The results showed that the center of pressure progression in the sagittal plane and knee momentum (i.e., residual thigh and prosthesis shank angular momentum about prosthetic knee joint) can potentially estimate the users’ perceptions of gait stability when experiencing disturbances. The findings of this study can help improve the development and evaluation of gait stability control algorithms in robotic prosthetic devices.}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Naseri, Amirreza and Lee, I-Chieh and Huang, He and Liu, Ming}, year={2023}, pages={4693–4702} }
 @article{fleming_liu_huang_2023, title={Neural prosthesis control restores near-normative neuromechanics in standing postural control}, volume={8}, ISSN={["2470-9476"]}, DOI={10.1126/scirobotics.adf5758}, abstractNote={Current lower-limb prostheses do not provide active assistance in postural control tasks to maintain the user’s balance, particularly in situations of perturbation. In this study, we aimed to address this missing function by enabling neural control of robotic lower-limb prostheses. Specifically, electromyographic (EMG) signals (amplified neural control signals) recorded from antagonistic residual ankle muscles were used to drive a robotic prosthetic ankle directly and continuously. Participants with transtibial amputation were recruited and trained in using the EMG-driven robotic ankle. We studied how using the EMG-controlled ankle affected the participants’ anticipatory and compensatory postural control strategies and stability under expected perturbations compared with using their daily passive devices. We investigated the similarity of neuromuscular coordination (by analyzing motor modules) of the participants, using either device in a postural sway task, to that of able-bodied controls. Results showed that, compared with their passive prosthesis, the EMG-controlled prosthesis enabled participants to use near-normative postural control strategies, as evidenced by improved between-limb symmetry in intact-prosthetic center-of-pressure and joint angle excursions. Participants substantially improved postural stability, as evidenced by a reduction in steps or falls using the EMG-controlled prosthetic ankle. Furthermore, after relearning to use residual ankle muscles to drive the robotic ankle in postural control, nearly all participants’ motor module structure shifted toward that observed in individuals without limb amputations. Here, we have demonstrated the potential benefit of direct EMG control of robotic lower limb prostheses to restore normative postural control strategies (both neural and biomechanical) toward enhancing standing postural stability in amputee users.}, number={83}, journal={SCIENCE ROBOTICS}, author={Fleming, Aaron and Liu, Wentao and Huang, He}, year={2023}, month={Oct} }
 @article{hinson_berman_lee_filer_huang_2023, title={Offline Evaluation Matters: Investigation of the Influence of Offline Performance of EMG-Based Neural-Machine Interfaces on User Adaptation, Cognitive Load, and Physical Efforts in a Real-Time Application}, volume={31}, ISSN={["1558-0210"]}, DOI={10.1109/TNSRE.2023.3297448}, abstractNote={There has been controversy about the value of offline evaluation of EMG-based neural-machine interfaces (NMIs) for their real-time application. Often, conclusions have been drawn after studying the correlation of the offline EMG decoding accuracy/error with the NMI user’s real-time task performance without further considering other important human performance metrics such as adaptation rate, cognitive load, and physical effort. To fill this gap, this study aimed to investigate the relationship between the offline decoding accuracy of EMG-based NMIs and user adaptation, cognitive load, and physical effort in real-time NMI use. Twelve non-disabled subjects participated in this study. For each subject, we established three EMG decoders that yielded different offline accuracy (low, moderate, and high) in predicting continuous hand and wrist motions. The subject then used each EMG decoder to perform a virtual hand posture matching task in real time with and without a secondary task as the evaluation trials. Results showed that the high-level offline performance decoders yield the fastest adaptation rate and highest posture matching completion rate with the least muscle effort in users during online testing. A secondary task increased the cognitive load and reduced real-time virtual task competition rate for all the decoders; however, the decoder with high offline accuracy still produced the highest task completion rate. These results imply that the offline performance of EMG-based NMIs provide important insight to users’ abilities to utilize them and should play an important role in research and development of novel NMI algorithms.}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Hinson, Robert M. and Berman, Joseph and Lee, I-Chieh and Filer, William G. and Huang, He}, year={2023}, pages={3055–3063} }
 @article{fisher_gaunt_huang_2023, title={Sensory restoration for improved motor control of prostheses}, volume={28}, ISSN={["2468-4511"]}, url={http://dx.doi.org/10.1016/j.cobme.2023.100498}, DOI={10.1016/j.cobme.2023.100498}, abstractNote={Somatosensory neuroprostheses are devices with the potential to restore the senses of touch and movement from prosthetic limbs for people with limb amputation or paralysis. By electrically stimulating the peripheral or central nervous system, these devices evoke sensations that appear to emanate from the missing or insensate limb, and when paired with sensors on the prosthesis, they can improve the functionality and embodiment of the prosthesis. There have been major advances in the design of these systems over the past decade, although several important steps remain before they can achieve widespread clinical adoption outside the lab setting. Here, we provide a brief overview of somatosensory neuroprostheses and explores these hurdles and potential next steps towards clinical translation.}, journal={CURRENT OPINION IN BIOMEDICAL ENGINEERING}, publisher={Elsevier BV}, author={Fisher, Lee E. and Gaunt, Robert A. and Huang, He}, year={2023}, month={Dec} }
 @article{hong_huang_2023, title={Towards Personalized Control for Powered Knee Prostheses: Continuous Impedance Functions and PCA-Based Tuning Method}, ISSN={["1945-7898"]}, DOI={10.1109/ICORR58425.2023.10304689}, abstractNote={Optimizing control parameters is crucial for personalizing prosthetic devices. The current method of finite state machine impedance control (FSM-IC) allows interaction with the user but requires time-consuming manual tuning. To improve efficiency, we propose a novel approach for tuning knee prostheses using continuous impedance functions (CIFs) and Principal Component Analysis (PCA). The CIFs, which represent stiffness, damping, and equilibrium angle, are modeled as fourth-order polynomials and optimized through convex optimization. By applying PCA to the CIFs, we extract principal components (PCs) that capture common features. The weights of these PCs serve as tuning parameters, allowing us to reconstruct various impedance functions. We validated this approach using data from 10 able-bodied individuals walking. The contributions of this study include: i) generating CIFs via convex optimization; ii) introducing a new tuning space based on the obtained CIFs; and iii) evaluating the feasibility of this tuning space.}, journal={2023 INTERNATIONAL CONFERENCE ON REHABILITATION ROBOTICS, ICORR}, author={Hong, Woolim and Huang, He}, year={2023} }
 @article{liu_wu_si_huang_2022, title={A New Robotic Knee Impedance Control Parameter Optimization Method Facilitated by Inverse Reinforcement Learning}, volume={7}, ISSN={["2377-3766"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85135753501&partnerID=MN8TOARS}, DOI={10.1109/LRA.2022.3194326}, abstractNote={Recent efforts in the design of intelligent controllers for configuring robotic prostheses have demonstrated new possibilities in improving mobility and restoring locomotion for individuals with lower-limb disabilities. In these efforts, personalizing the controller of the robotic device is a crucial step in order to meet individual user's needs and physical conditions. Reinforcement learning (RL) based control designs are among some of the most promising approaches to achieving real-time, optimal adaptive tuning capability. However, such designs to date rely on subjectively determining human-robot walking performance measures, commonly in a quadratic form. To further automate the RL design for robotic knee control parameter tuning and potentially improve human-robot locomotion performance, this study introduces a new bilevel optimization method to objectively specify such control design performance measures via inverse reinforcement learning (IRL), which in turn, will be used in low level (forward) RL design of the impedance control parameters. We demonstrate the effectiveness of the bilevel optimization approach with improved human-robot walking performance using systematic OpenSim simulation studies.}, number={4}, journal={IEEE ROBOTICS AND AUTOMATION LETTERS}, author={Liu, Wentao and Wu, Ruofan and Si, Jennie and Huang, He}, year={2022}, month={Oct}, pages={10882–10889} }
 @article{liu_naseri_lee_hu_lewek_huang_2023, title={A simplified model for whole-body angular momentum calculation}, volume={111}, ISSN={["1873-4030"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85144824437&partnerID=MN8TOARS}, DOI={10.1016/j.medengphy.2022.103944}, abstractNote={The capability to monitor gait stability during everyday life could provide key information to guide clinical intervention to patients with lower limb disabilities. Whole body angular momentum (Lbody) is a convenient stability indicator for wearable motion capture systems. However, Lbody is costly to estimate, because it requires monitoring all major body segment using expensive sensor elements. In this study, we developed a simplified rigid body model by merging connected body segments to reduce the number of body segments, which need to be monitored. We demonstrated that the Lbody could be estimated by a seven-segment model accurately for both people with and without lower extremity amputation.}, journal={MEDICAL ENGINEERING & PHYSICS}, author={Liu, Ming and Naseri, Amirreza and Lee, I-Chieh and Hu, Xiaogang and Lewek, Michael D. and Huang, He}, year={2023}, month={Jan} }
 @inproceedings{nalam_tu_li_si_huang_2022, title={Admittance Control Based Human-in-the-Loop Optimization for Hip Exoskeleton Reduces Human Exertion during Walking}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85136333931&partnerID=MN8TOARS}, DOI={10.1109/ICRA46639.2022.9811553}, abstractNote={Human-in-the-loop (HIL) optimization usually optimizes assistive torque of exoskeletons to minimize the human's energetic expenditure in walking, quantified by metabolic cost. This formulation can, however, result in altered gait pattern of the human joint from the natural pattern, which is undesired. In this paper, we proposed a novel concept of HIL optimization of a hip exoskeleton. The optimization goal was to maintain the hip kinematics while providing optimal mechanical energy from the exoskeleton by modulating the admittance control. Policy iteration was used to optimize the switching time within the gait phase, at which a single parameter of the admittance controller was altered to provide assistance. The stiffness and equilibrium angle were considered as the two parameters for altering at the switching time, resulting in three possible modes of operation for the algorithm: (i) switching the equilibrium point, (ii) switching stiffness while equilibrium point is set at maximum extension and, (iii) maximum flexion. The optimization algorithm was found to converge for all three modes, with the equilibrium mode resulting in multiple solutions. Further analysis of power injected by the exoskeleton in the three modes showed that the first and third mode reduced human energetic exertion while the second mode increased human exertion. Implications of the results as well as the observed muscle activation patterns in response to assistance are discussed.}, booktitle={Proceedings - IEEE International Conference on Robotics and Automation}, author={Nalam, V. and Tu, X. and Li, M. and Si, J. and Huang, H.H.}, year={2022}, pages={6743–6749} }
 @article{naseri_liu_lee_liu_huang_2022, title={Characterizing Prosthesis Control Fault During Human-Prosthesis Interactive Walking Using Intrinsic Sensors}, volume={7}, ISSN={["2377-3766"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85133795882&partnerID=MN8TOARS}, DOI={10.1109/LRA.2022.3186503}, abstractNote={The physical interactions between wearable lower limb robots and humans have been investigated to inform effective robot design for walking augmentation. However, human-robot interactions when internal faults occur within robots have not been systematically reported, but it is essential to improve the robustness of robotic devices and ensure the user’s safety. This letter aims to (1) present a methodology to characterize the behavior of the robotic transfemoral prosthesis as an effective wearable robot platform while interacting with the users in the presence of internal faults, and (2) identify the potential data sources for accurate detection of the prosthesis fault. We first obtained the human perceived response in terms of their walking stability when the prosthesis control fault (inappropriate intrinsic control output/command) was emulated/applied in level-ground walking. Then the measurements and their features, obtained from the transfemoral prosthesis, were examined for the emulated faults that elicited a sense of instability in human users. The optimal features that contributed the most in separating faulty interaction from the normal walking condition were determined using two machine-learning-based approaches: One-Class Support Vector Machine (OCSVM) and Mahalanobis Distance (MD) classifier. The OCSVM anomaly detector could achieve an average sensitivity of 85.7% and an average false alarm rate of 1.7% with a reasonable detecting time of 147.6 ms for detecting emulated control errors among all subjects. The result demonstrates the potential of using machine-learning-based schemes in identifying prosthesis control faults based on intrinsic sensors on the prosthesis. This study presents a procedure to study human-robot fault tolerance and inform the future design of robust prosthesis control.}, number={3}, journal={IEEE ROBOTICS AND AUTOMATION LETTERS}, author={Naseri, Amirreza and Liu, Ming and Lee, I-Chieh and Liu, Wentao and Huang, He}, year={2022}, month={Jul}, pages={8307–8314} }
 @inproceedings{park_berman_dodson_liu_matthew_huang_kaber_ruiz_zahabi_2022, title={Cognitive Workload Classification of Upper-limb Prosthetic Devices}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85146271387&partnerID=MN8TOARS}, DOI={10.1109/ICHMS56717.2022.9980676}, abstractNote={Limb amputation can cause severe functional disability in performing activities of daily living (ADLs). Using prosthetic devices as aids for such activities requires substantial cognitive resources. Machine Learning (ML) algorithms can be used to predict cognitive workload (CW) of prosthetic device prototypes early in the design process and serve as a tool for improving device usability. The objective of this study was to explore subsets of input features that can be easily captured during early stages of the design cycle to classify CW of electromyography (EMG)-based upper-limb prostheses. An experiment was conducted with 30 participants to collect task performance and pupillometry data, and to provide a basis for generating cognitive performance model (CPM) outcomes. Three ML algorithms, including the random forest (RF), support vector machine (SVM), and naive Bayesian (NB) classifier were developed. The most important subset of features was selected based on classification accuracy and computational and experimental cost. Findings revealed that the CPM outcomes and prosthetic device configuration were the most important features for reasonably classifying CW responses under low cost. Also, the SVM classifier can be used for near-real time classification of CW. Future studies should include additional data and improve hyperparameter tuning parameters, as well as advanced CPM techniques to improve the performance of algorithms.}, booktitle={Proceedings of the 2022 IEEE International Conference on Human-Machine Systems, ICHMS 2022}, author={Park, J. and Berman, J. and Dodson, A. and Liu, Y. and Matthew, A. and Huang, H. and Kaber, D. and Ruiz, J. and Zahabi, M.}, year={2022} }
 @article{shah_fleming_nalam_liu_huang_2022, title={Design of EMG-driven Musculoskeletal Model for Volitional Control of a Robotic Ankle Prosthesis}, volume={2022-October}, ISSN={["2153-0858"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85146352781&partnerID=MN8TOARS}, DOI={10.1109/IROS47612.2022.9981305}, abstractNote={Existing robotic lower-limb prostheses use autonomous control to address cyclic, locomotive tasks, but are inadequate in adapting to variations in non-cyclic and unpredictable tasks. This study aims to address this challenge by designing a novel electromyography (EMG)-driven musculoskeletal model for volitional control of a robotic ankle-foot prosthesis. The proposed controller ensures continuous control of the device, allowing users to freely manipulate the prosthesis behavior. A Hill-type muscle model was implemented to model a dorsiflexor and a plantarflexor to function around a virtual ankle joint. The model parameters for a subject specific model was determined by fitting the model to the experimental data collected from an able-bodied subject. EMG signals recorded from antagonist muscle pairs were used to activate the virtual muscle models. This model-based approach was then validated via offline simulations and real-time prosthesis control. Additionally, the feasibility of the proposed prosthesis control on assisting the user's functional tasks was demonstrated. The present control may further improve the function of robotic prosthesis for supporting versatile activities in individuals with lower-limb amputations.}, journal={2022 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS)}, author={Shah, Chinmay and Fleming, Aaron and Nalam, Varun and Liu, Ming and Huang, He}, year={2022}, pages={12261–12266} }
 @article{shah_fleming_nalam_huang_2022, title={Design of EMG-driven musculoskeletal model for volitional control of a robotic ankle prosthesis}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85125688202&partnerID=MN8TOARS}, journal={arXiv}, author={Shah, C. and Fleming, A. and Nalam, V. and Huang, H.}, year={2022} }
 @article{zhong_silva_tran_huang_lobaton_2021, title={Efficient Environmental Context Prediction for Lower Limb Prostheses}, volume={52}, ISSN={["2168-2232"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85111025519&partnerID=MN8TOARS}, DOI={10.1109/TSMC.2021.3084036}, abstractNote={Environmental context prediction is important for wearable robotic applications, such as terrain-adaptive control. System efficiency is critical for wearable robots, in which system resources (e.g., processors and memory) are highly constrained. This article aims to address the system efficiency of real-time environmental context prediction for lower limb prostheses. First, we develop an uncertainty-aware frame selection strategy that can dynamically select frames according to lower limb motion and uncertainty captured by Bayesian neural networks (BNNs) for environment prediction. We further propose a dynamic Bayesian gated recurrent unit (D-BGRU) network to address the inconsistent frame rate which is a side effect of the dynamic frame selection. Second, we investigate the effects on the tradeoff between computational complexity and environment prediction accuracy of adding additional sensing modalities (e.g., GPS and an on-glasses camera) into the system. Finally, we implement and optimize our framework for embedded hardware, and evaluate the real-time inference accuracy and efficiency of classifying six types of terrains. The experiments show that our proposed frame selection strategy can reduce more than 90% of the computations without sacrificing environment prediction accuracy, and can be easily extended to the situation of multimodality fusion. We achieve around 93% prediction accuracy with less than one frame to be processed per second. Our model has 6.4 million 16-bit float numbers and takes 44 ms to process each frame on a lightweight embedded platform (NVIDIA Jetson TX2).}, number={6}, journal={IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Zhong, Boxuan and Silva, Rafael Luiz and Tran, Michael and Huang, He and Lobaton, Edgar}, year={2021}, month={Jun} }
 @article{vargas_huang_zhu_hu_2022, title={Evoked Tactile Feedback and Control Scheme on Functional Utility of Prosthetic Hand}, volume={7}, ISSN={["2377-3766"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85122303364&partnerID=MN8TOARS}, DOI={10.1109/LRA.2021.3139147}, abstractNote={Fine manual control relies on intricate action-perception coupling to effectively interact with objects. Here, we evaluated how electrically evoked artificial tactile sensation can be integrated into the functional utility of a prosthetic hand. Using different myoelectric-control strategies, participants performed a modified box-and-block task using a prosthetic hand. Transcutaneous nerve stimulation was employed to elicit somatotopic fingertip tactile feedback reflecting prosthetic fingertip forces. This feedback was evoked using an electrode grid placed along the participants’ upper arm targeting the median and ulnar nerve bundles. Myoelectric signals from the finger flexor and extensor controlled the prosthetic joint velocity or position. Participants lifted, held, and transported cubes of varying weights using their minimum grip forces. The results showed that participants exerted lower forces and presented lower number of failed trials (prematurely dropped objects) when feedback was provided with respect to without feedback. We also found that position control required more flexor muscle activation compared with velocity control when tactile feedback was provided. Our findings reveal that non-invasively evoked tactile feedback could be used to effectively enable human-in-the-loop control of a prosthetic hand. The outcomes can provide a platform to characterize the action-perception couplings during prosthetic control, in order to improve user experience and system functionality.}, number={2}, journal={IEEE ROBOTICS AND AUTOMATION LETTERS}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Vargas, Luis and Huang, He and Zhu, Yong and Hu, Xiaogang}, year={2022}, month={Apr}, pages={1300–1307} }
 @article{rubin_zheng_huang_hu_2022, title={Finger Force Estimation Using Motor Unit Discharges Across Forearm Postures}, volume={69}, ISSN={["1558-2531"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85125331087&partnerID=MN8TOARS}, DOI={10.1109/TBME.2022.3153448}, abstractNote={Background: Myoelectric- based decoding has gained popularity in upper- limb neural-machine interfaces. Motor unit (MU) firings decomposed from surface electromyographic (EMG) signals can represent motor intent, but EMG properties at different arm configurations can change due to electrode shift and differing neuromuscular states. This study investigated whether isometric fingertip force estimation using MU firings is robust to forearm rotations from a neutral to either a fully pronated or supinated posture. Methods: We extracted MU information from high- density EMG of the extensor digitorum communis in two ways: (1) Decomposed EMG in all three postures (MU-AllPost); and (2) Decomposed EMG in neutral posture (MU-Neu), and extracted MUs (separation matrix) were applied to other postures. Populational MU firing frequency estimated forces scaled to subjects’ maximum voluntary contraction (MVC) using a regression analysis. The results were compared with the conventional EMG-amplitude method. Results: We found largely similar root-mean-square errors (RMSE) for the two MU-methods, indicating that MU decomposition was robust to postural differences. MU-methods demonstrated lower RMSE in the ring (EMG = 6.23, MU-AllPost = 5.72, MU-Neu = 5.64% MVC) and pinky (EMG = 6.12, MU-AllPost = 4.95, MU-Neu = 5.36% MVC) fingers, with mixed results in the middle finger (EMG = 5.47, MU-AllPost = 5.52, MU-Neu = 6.19% MVC). Conclusion: Our results suggest that MU firings can be extracted reliably with little influence from forearm posture, highlighting its potential as an alternative decoding scheme for robust and continuous control of assistive devices.}, number={9}, journal={IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING}, author={Rubin, Noah and Zheng, Yang and Huang, He and Hu, Xiaogang}, year={2022}, month={Sep}, pages={2767–2775} }
 @article{li_zhong_lobaton_huang_2022, title={Fusion of Human Gaze and Machine Vision for Predicting Intended Locomotion Mode}, volume={30}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85128628295&partnerID=MN8TOARS}, DOI={10.1109/TNSRE.2022.3168796}, abstractNote={Predicting the user’s intended locomotion mode is critical for wearable robot control to assist the user’s seamless transitions when walking on changing terrains. Although machine vision has recently proven to be a promising tool in identifying upcoming terrains in the travel path, existing approaches are limited to environment perception rather than human intent recognition that is essential for coordinated wearable robot operation. Hence, in this study, we aim to develop a novel system that fuses the human gaze (representing user intent) and machine vision (capturing environmental information) for accurate prediction of the user’s locomotion mode. The system possesses multimodal visual information and recognizes user’s locomotion intent in a complex scene, where multiple terrains are present. Additionally, based on the dynamic time warping algorithm, a fusion strategy was developed to align temporal predictions from individual modalities while producing flexible decisions on the timing of locomotion mode transition for wearable robot control. System performance was validated using experimental data collected from five participants, showing high accuracy (over 96% in average) of intent recognition and reliable decision-making on locomotion transition with adjustable lead time. The promising results demonstrate the potential of fusing human gaze and machine vision for locomotion intent recognition of lower limb wearable robots.}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Li, Minhan and Zhong, Boxuan and Lobaton, Edgar and Huang, He}, year={2022}, pages={1103–1112} }
 @article{fylstra_lee_li_lewek_huang_2022, title={Human-Prosthesis Cooperation: Combining Adaptive Prosthesis Control with Visual Feedback Guided Gait}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85134127407&partnerID=MN8TOARS}, DOI={10.21203/rs.3.rs-1670416}, journal={ResearchSquare}, author={Fylstra, B.L. and Lee, I.C. and Li, M. and Lewek, M. and Huang, H.}, year={2022} }
 @article{fylstra_lee_li_lewek_huang_2022, title={Human-prosthesis cooperation: combining adaptive prosthesis control with visual feedback guided gait}, volume={19}, ISSN={["1743-0003"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85144315570&partnerID=MN8TOARS}, DOI={10.1186/s12984-022-01118-z}, abstractNote={Abstract
                Background
                Personalizing prosthesis control is often structured as human-in-the-loop optimization. However, gait performance is influenced by both human control and intelligent prosthesis control. Hence, we need to consider both human and prosthesis control, and their cooperation, to achieve desired gait patterns. In this study, we developed a novel paradigm that engages human gait control via user-fed visual feedback (FB) of stance time to cooperate with automatic prosthesis control tuning. Three initial questions were studied: (1) does user control of gait timing (via visual FB) help the prosthesis tuning algorithm to converge faster? (2) in turn, does the prosthesis control influence the user’s ability to reach and maintain the target stance time defined by the feedback? and (3) does the prosthesis control parameters tuned with extended stance time on prosthesis side allow the user to maintain this potentially beneficial behavior even after feedback is removed (short- and long-term retention)?
              
                Methods
                A reinforcement learning algorithm was used to achieve prosthesis control to meet normative knee kinematics in walking. A visual FB system cued the user to control prosthesis-side stance time to facilitate the prosthesis tuning goal. Seven individuals without amputation (AB) and four individuals with transfemoral amputation (TFA) walked with a powered knee prosthesis on a treadmill. Participants completed prosthesis auto-tuning with three visual feedback conditions: no FB, self-selected stance time FB (SS FB), and increased stance time FB (Inc FB). The retention of FB effects was studied by comparing the gait performance across three different prosthesis controls, tuned with different visual FB.
              
                Results
                (1) Human control of gait timing reduced the tuning duration in individuals without amputation, but not for individuals with TFA. (2) The change of prosthesis control did not influence users’ ability to reach and maintain the visual FB goal. (3) All participants increased their prosthesis-side stance time with the feedback and maintain it right after feedback was removed. However, in the post-test, the prosthesis control parameters tuned with visual FB only supported a few participants with longer stance time and better stance time symmetry.
              
                Conclusions
                The study provides novel insights on human-prosthesis interaction when cooperating in walking, which may guide the future successful adoption of this paradigm in prosthesis control personalization or human-in-the-loop optimization to improve the prosthesis user’s gait performance.
              }, number={1}, journal={JOURNAL OF NEUROENGINEERING AND REHABILITATION}, author={Fylstra, Bretta L. and Lee, I-Chieh and Li, Minhan and Lewek, Michael D. and Huang, He}, year={2022}, month={Dec} }
 @article{zhang_nalam_tu_li_si_lewek_huang_2022, title={Imposing Healthy Hip Motion Pattern and Range by Exoskeleton Control for Individualized Assistance}, volume={7}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85135761860&partnerID=MN8TOARS}, DOI={10.1109/LRA.2022.3196105}, abstractNote={Powered exoskeletons are promising devices to improve the walking patterns of people with neurological impairments. Providing personalized external assistance though is challenging due to uncertainties and the time-varying nature of human-robot interaction. Recently, human-in-the-loop (HIL) optimization has been investigated for providing assistance to minimize energetic expenditure, usually quantified by metabolic cost. However, this full-body global effect evaluation may not directly reflect the local functions of the targeted joint(s). This makes it difficult to assess the direct effect when robotic assistance is provided. In addition, the HIL optimization method usually does not take into account local joint trajectories, a consideration that is important in imposing healthy joint movements and gait patterns for individuals with lower limb motor deficits. In this paper, we propose a model-free reinforcement learning (RL)-based control framework to achieve a normative range of motion and gait pattern of the hip joint during walking. Our RL-based control provides personalized assistance torque profile by heuristically manipulating three control parameters for hip flexion and extension, respectively, during walking. A least square policy iteration was devised to optimize a cost function associated with control efforts and hip joint trajectory errors by tuning the control parameters. To evaluate the performance of the design approach, a compression sleeve was used to constrain the hip joint of unimpaired human participants to simulate motor deficits. The proposed RL control successfully achieved the desired goal of enlarging the hip joint's range of motion in three participants walking on a treadmill.}, number={4}, journal={IEEE Robotics and Automation Letters}, author={Zhang, Qiang and Nalam, Varun and Tu, Xikai and Li, Minhan and Si, Jennie and Lewek, Michael D. and Huang, He Helen}, year={2022}, pages={11126–11133} }
 @article{liu_zhong_wu_fylstra_si_huang_2022, title={Inferring Human-Robot Performance Objectives During Locomotion Using Inverse Reinforcement Learning and Inverse Optimal Control}, volume={7}, ISSN={["2377-3766"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85123342480&partnerID=MN8TOARS}, DOI={10.1109/LRA.2022.3143579}, abstractNote={Quantitatively characterizing a locomotion performance objective for a human-robot system is an important consideration in the assistive wearable robot design towards human-robot symbiosis. This problem, however, has only been addressed sparsely in the literature. In this study, we propose a new inverse approach from observed human-robot walking behavior to infer a human-robot collective performance objective represented in a quadratic form. By an innovative design of human experiments and simulation study, respectively, we validated the effectiveness of two solution approaches to solving the inverse problem using inverse reinforcement learning (IRL) and inverse optimal control (IOC). The IRL-based experiments of human walking with robotic transfemoral prosthesis validated the realistic applicability of the proposed inverse approach, while the IOC-based analysis provided important human-robot system properties such as stability and robustness that are difficult to obtain from human experiments. This study introduces a new tool to the field of wearable lower limb robots. It is expected to be expandable to quantify joint human-robot locomotion performance objectives for personalizing wearable robot control in the future.}, number={2}, journal={IEEE ROBOTICS AND AUTOMATION LETTERS}, author={Liu, Wentao and Zhong, Junmin and Wu, Ruofan and Fylstra, Bretta L. and Si, Jennie and Huang, He}, year={2022}, month={Apr}, pages={2549–2556} }
 @article{moreno_vitiello_walsh_huang_mohammed_2022, title={Introduction to the Special Section on Wearable Robots}, volume={38}, ISSN={["1941-0468"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85132402999&partnerID=MN8TOARS}, DOI={10.1109/TRO.2022.3176744}, abstractNote={The papers in this special section focus on the development and applications supported by wearable robots. Wearable powered robots may be used for functional substitution in patients suffering from motor disorders, rehabilitation, assistance, and strength augmentation. Despite recent technological and scientific achievements, more research is needed to realize the promise of intuitive, easy-to-wear, safe, and effective wearable robots}, number={3}, journal={IEEE TRANSACTIONS ON ROBOTICS}, author={Moreno, Juan C. and Vitiello, Nicola and Walsh, Conor and Huang, He and Mohammed, Samer}, year={2022}, month={Jun}, pages={1338–1342} }
 @article{lee_fylstra_liu_lenzi_huang_2022, title={Is there a trade-off between economy and task goal variability in transfemoral amputee gait?}, volume={19}, ISSN={["1743-0003"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85126546016&partnerID=MN8TOARS}, DOI={10.1186/s12984-022-01004-8}, abstractNote={Abstract
                Background
                Energy cost minimization has been widely accepted to regulate gait. Optimization principles have been frequently used to explain how individuals adapt their gait pattern. However, there have been rare attempts to account for the role of variability in this optimization process. Motor redundancy can enable individuals to perform tasks reliably while achieving energy optimization. However, we do not know how the non-goal-equivalent and goal-equivalent variability is regulated. In this study, we investigated how unilateral transfemoral amputees regulate step and stride variability based on the task to achieve energy economy.
              
                Methods
                Nine individuals with unilateral transfemoral amputation walked on a treadmill at speeds of 0.6, 0.8, 1.0, 1.2 and 1.4 m/s using their prescribed passive prostheses. We calculated the step-to-step and stride-to-stride variability and applied goal equivalent manifold (GEM) based control to decompose goal-equivalent and non-goal-equivalent manifold. To quantify the energy economy, the energy recovery rate (R) was calculated based on potential energy and kinetic energy. Comparisons were made between GEM variabilities and commonly used standard deviation measurements. A linear regression model was used to investigate the trade-off between R and GEM variabilities.
              
                Results
                Our analysis shows greater variability along the goal-equivalent manifold compared to the non-goal-equivalent manifold (p < 0.001). Moreover, our analysis shows lower energy recovery rate for amputee gait compared to nonamputee gait (at least 20% less at faster walking speed). We found a negative relationship between energy recovery rate and non-goal-equivalent variability. Compared to the standard deviation measurements, the variability decomposed using GEM reflected the preferred walking speed and the limitation of the passive prosthetic device.
              
                Conclusion
                Individuals with amputation cleverly leverage task redundancy, regulating step and stride variability to the GEM. This result suggests that task redundancy enables unilateral amputees to benefit from motor variability in terms of energy economy. The differences observed between prosthetic step and intact step support the development of prosthetic limbs capable of enhancing positive work during the double support phase and of powered prosthesis controllers that allow for variability along the task space while minimizing variability that interferes with the task goal. This study provides a different perspective on amputee gait analysis and challenges the field to think differently about the role of variability.
              }, number={1}, journal={JOURNAL OF NEUROENGINEERING AND REHABILITATION}, author={Lee, I-Chieh and Fylstra, Bretta L. and Liu, Ming and Lenzi, Tommaso and Huang, He}, year={2022}, month={Mar} }
 @article{hinson_berman_filer_kamper_hu_huang_2023, title={Offline Evaluation Matters: Investigation of the Influence of Offline Performance on Real-Time Operation of Electromyography-Based Neural-Machine Interfaces}, volume={31}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85144812628&partnerID=MN8TOARS}, DOI={10.1109/TNSRE.2022.3226229}, abstractNote={There has been a debate on the most appropriate way to evaluate electromyography (EMG)-based neural-machine interfaces (NMIs). Accordingly, this study examined whether a relationship between offline kinematic predictive accuracy (R2) and user real-time task performance while using the interface could be identified. A virtual posture-matching task was developed to evaluate motion capture-based control and myoelectric control with artificial neural networks (ANNs) trained to low (R2 ≈ 0.4), moderate (R2 ≈ 0.6), and high ( $\text {R}^{\vphantom {\text {D}^{\text {a}}}{2}} \approx 0.8$ ) offline performance levels. Twelve non-disabled subjects trained with each offline performance level decoder before evaluating final real-time posture matching performance. Moderate to strong relationships were detected between offline performance and all real-time task performance metrics: task completion percentage (r = 0.66, p < 0.001), normalized task completion time (r = −0.51, p = 0.001), path efficiency (r = 0.74, p < 0.001), and target overshoots (r = −0.79, p < 0.001). Significant improvements in each real-time task evaluation metric were also observed between the different offline performance levels. Additionally, subjects rated myoelectric controllers with higher offline performance more favorably. The results of this study support the use and validity of offline analyses for optimization of NMIs in myoelectric control research and development.}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Hinson, Robert M. and Berman, Joseph and Filer, William and Kamper, Derek and Hu, Xiaogang and Huang, He}, year={2023}, pages={680–689} }
 @article{gao_si_wen_li_huang_2021, title={Reinforcement Learning Control of Robotic Knee With Human-in-the-Loop by Flexible Policy Iteration}, volume={33}, ISSN={["2162-2388"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85105845147&partnerID=MN8TOARS}, DOI={10.1109/TNNLS.2021.3071727}, abstractNote={We are motivated by the real challenges presented in a human–robot system to develop new designs that are efficient at data level and with performance guarantees, such as stability and optimality at system level. Existing approximate/adaptive dynamic programming (ADP) results that consider system performance theoretically are not readily providing practically useful learning control algorithms for this problem, and reinforcement learning (RL) algorithms that address the issue of data efficiency usually do not have performance guarantees for the controlled system. This study fills these important voids by introducing innovative features to the policy iteration algorithm. We introduce flexible policy iteration (FPI), which can flexibly and organically integrate experience replay and supplemental values from prior experience into the RL controller. We show system-level performances, including convergence of the approximate value function, (sub)optimality of the solution, and stability of the system. We demonstrate the effectiveness of the FPI via realistic simulations of the human–robot system. It is noted that the problem we face in this study may be difficult to address by design methods based on classical control theory as it is nearly impossible to obtain a customized mathematical model of a human–robot system either online or offline. The results we have obtained also indicate the great potential of RL control to solving realistic and challenging problems with high-dimensional control inputs.}, number={10}, journal={IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Gao, Xiang and Si, Jennie and Wen, Yue and Li, Minhan and Huang, He}, year={2021}, month={May} }
 @article{wu_li_yao_liu_si_huang_2022, title={Reinforcement Learning Impedance Control of a Robotic Prosthesis to Coordinate With Human Intact Knee Motion}, volume={7}, ISSN={["2377-3766"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85131742245&partnerID=MN8TOARS}, DOI={10.1109/LRA.2022.3179420}, abstractNote={This study aims to demonstrate reinforcement learning tracking control for automatically configuring the impedance parameters of a robotic knee prosthesis. While our previous studies involving human subjects have focused on tuning the impedance control parameters to meet a fixed, subjectively prescribed target motion profile to enable continuous walking with human-in-the-loop, in this paper we develop a new tracking control solution for a robotic knee to mimic the motion of the intact knee. As such, we replaced the prescribed target knee motion by an automatically generated profile based on the intact knee. As the profile of the intact knee varies over time due to human adaptation, we are presented with a challenging tracking control problem in the context of classical control theory. By formulating the “echo control” of the robotic knee as a reinforcement learning problem, we provide a promising new tool for real-time tracking control design without explicitly representing the underlying dynamics using a mathematical model, which can be difficult to obtain for a human-robot system. Additionally, our results may inspire future studies and new robotic prosthesis impedance control designs that can potentially coordinate between the intact and the robotic limbs toward daily use of the robotic device.}, number={3}, journal={IEEE ROBOTICS AND AUTOMATION LETTERS}, author={Wu, Ruofan and Li, Minhan and Yao, Zhikai and Liu, Wentao and Si, Jennie and Huang, He}, year={2022}, month={Jul}, pages={7014–7020} }
 @article{zhong_huang_lobaton_2022, title={Reliable Vision-Based Grasping Target Recognition for Upper Limb Prostheses}, volume={52}, ISSN={["2168-2275"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85126389026&partnerID=MN8TOARS}, DOI={10.1109/TCYB.2020.2996960}, abstractNote={Computer vision has shown promising potential in wearable robotics applications (e.g., human grasping target prediction and context understanding). However, in practice, the performance of computer vision algorithms is challenged by insufficient or biased training, observation noise, cluttered background, etc. By leveraging Bayesian deep learning (BDL), we have developed a novel, reliable vision-based framework to assist upper limb prosthesis grasping during arm reaching. This framework can measure different types of uncertainties from the model and data for grasping target recognition in realistic and challenging scenarios. A probability calibration network was developed to fuse the uncertainty measures into one calibrated probability for online decision making. We formulated the problem as the prediction of grasping target while arm reaching. Specifically, we developed a 3-D simulation platform to simulate and analyze the performance of vision algorithms under several common challenging scenarios in practice. In addition, we integrated our approach into a shared control framework of a prosthetic arm and demonstrated its potential at assisting human participants with fluent target reaching and grasping tasks.}, number={3}, journal={IEEE TRANSACTIONS ON CYBERNETICS}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Zhong, Boxuan and Huang, He and Lobaton, Edgar}, year={2022}, month={Mar}, pages={1750–1762} }
 @article{vargas_huang_zhu_kamper_hu_2022, title={Resembled Tactile Feedback for Object Recognition Using a Prosthetic Hand}, volume={7}, ISSN={["2377-3766"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85136090957&partnerID=MN8TOARS}, DOI={10.1109/LRA.2022.3196958}, abstractNote={Tactile feedback in the hand is essential for interaction with objects. Here, we evaluated how artificial tactile sensation affected the recognition of object properties using a myoelectrically controlled prosthetic hand. Electromyogram signals from the flexor and extensor finger muscles were used to continuously control either prosthetic joint velocity or position. Participants grasped objects of varying shape or size using the prosthetic hand. Tactile feedback was evoked by transcutaneous nerve stimulation along the participant's upper arm and modulated based on the prosthetic-object contact force. Multi-channel electrical stimulation targeted the median and ulnar nerve bundles to produce resembled tactile sensations at distinct hand regions. The results showed that participants could gauge the onset timing of tactile feedback to discern object shape and size. We also found that the position-controller led to a greater recognition accuracy of object size compared with velocity-control, potentially due to supplemental joint position information from muscle activation level. Our findings demonstrate that non-invasive tactile feedback can enable effective object shape and size recognition during prosthetic control. The evaluation of tactile feedback across myoelectric controllers can help understand the interplay between sensory and motor pathways involved in the control of assistive devices.}, number={4}, journal={IEEE ROBOTICS AND AUTOMATION LETTERS}, author={Vargas, Luis and Huang, He and Zhu, Yong and Kamper, Derek and Hu, Xiaogang}, year={2022}, month={Oct}, pages={10977–10984} }
 @article{wu_yao_si_huang_2022, title={Robotic Knee Tracking Control to Mimic the Intact Human Knee Profile Based on Actor-Critic Reinforcement Learning}, volume={9}, ISSN={["2329-9274"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85118313741&partnerID=MN8TOARS}, DOI={10.1109/JAS.2021.1004272}, abstractNote={We address a state-of-the-art reinforcement learning (RL) control approach to automatically configure robotic prosthesis impedance parameters to enable end-to-end, continuous locomotion intended for transfemoral amputee subjects. Specifically, our actor-critic based RL provides tracking control of a robotic knee prosthesis to mimic the intact knee profile, This is a significant advance from our previous RL based automatic tuning of prosthesis control parameters which have centered on regulation control with a designer prescribed robotic knee profile as the target. In addition to presenting the tracking control algorithm based on direct heuristic dynamic programming (dHDP), we provide a control performance guarantee including the case of constrained inputs. We show that our proposed tracking control possesses several important properties, such as weight convergence of the learning networks, Bellman (sub) optimality of the cost-to-go value function and control input, and practical stability of the human-robot system. We further provide a systematic simulation of the proposed tracking control using a realistic human-robot system simulator, the OpenSim, to emulate how the dHDP enables level ground walking, walking on different terrains and at different paces. These results show that our proposed dHDP based tracking control is not only theoretically suitable, but also practically useful.}, number={1}, journal={IEEE-CAA JOURNAL OF AUTOMATICA SINICA}, author={Wu, Ruofan and Yao, Zhikai and Si, Jennie and Huang, He Helen}, year={2022}, month={Jan}, pages={19–30} }
 @article{hinson jr_saul_kamper_huang_2022, title={Sensitivity analysis guided improvement of an electromyogram-driven lumped parameter musculoskeletal hand model}, volume={141}, ISSN={["1873-2380"]}, url={http://dx.doi.org/10.1016/j.jbiomech.2022.111200}, DOI={10.1016/j.jbiomech.2022.111200}, abstractNote={EMG-driven neuromusculoskeletal models have been used to study many impairments and hold great potential to facilitate human–machine interactions for rehabilitation. A challenge to successful clinical application is the need to optimize the model parameters to produce accurate kinematic predictions. In order to identify the key parameters, we used Monte-Carlo simulations to evaluate the sensitivities of wrist and metacarpophalangeal (MCP) flexion/extension prediction accuracies for an EMG-driven, lumped-parameter musculoskeletal model. Four muscles were modeled with 22 total optimizable parameters. Model predictions from EMG were compared with measured joint angles from 11 able-bodied subjects. While sensitivities varied by muscle, we determined muscle moment arms, maximum isometric force, and tendon slack length were highly influential, while passive stiffness and optimal fiber length were less influential. Removing the two least influential parameters from each muscle reduced the optimization search space from 22 to 14 parameters without significantly impacting prediction correlation (wrist: 0.90 ± 0.05 vs 0.90 ± 0.05, p = 0.96; MCP: 0.74 ± 0.20 vs 0.70 ± 0.23, p = 0.51) and normalized root mean square error (wrist: 0.18 ± 0.03 vs 0.19 ± 0.03, p = 0.16; MCP: 0.18 ± 0.06 vs 0.19 ± 0.06, p = 0.60). Additionally, we showed that wrist kinematic predictions were insensitive to parameters of the modeled MCP muscles. This allowed us to develop a novel optimization strategy that more reliably identified the optimal set of parameters for each subject (27.3 ± 19.5%) compared to the baseline optimization strategy (6.4 ± 8.1%; p = 0.004). This study demonstrated how sensitivity analyses can be used to guide model refinement and inform novel and improved optimization strategies, facilitating implementation of musculoskeletal models for clinical applications.}, journal={JOURNAL OF BIOMECHANICS}, publisher={Elsevier BV}, author={Hinson Jr, Robert Jr and Saul, Katherine and Kamper, Derek and Huang, He}, year={2022}, month={Aug} }
 @article{yuan_bai_driscoll_liu_huang_feng_2022, title={Standing and Walking Attention Visual Field (SWAVF) task: A new method to assess visuospatial attention during walking}, volume={104}, ISSN={["1872-9126"]}, url={http://dx.doi.org/10.1016/j.apergo.2022.103804}, DOI={10.1016/j.apergo.2022.103804}, abstractNote={Visuospatial attention during walking has been associated with pedestrian safety and fall risks. However, visuospatial attention measures during walking remained under-explored. Current studies introduced a newly-developed Standing and Walking Visual Attention Field (SWAVF) task to assess visuospatial attention during walking and examined its reliability, validity, and stability. Thirty young adults completed a traditional computerized Attention Visual Field (AVF) task while sitting, and the SWAVF task under walking and standing settings. Nine participants also performed the SWAVF task under additional distraction conditions. Results showed good split-half reliability during standing (r = 0.70) and walking (r = 0.69), moderate concurrent validity with the sitting AVF task (r = 0.42), moderate convergent validity between the standing and walking settings (r = 0.69), good construct validity, and moderate rank-order stability (r = 0.53). Overall, the SWAVF task showed good psychometric properties. Potential applications to the evaluation of prosthetic and other exoskeleton devices, smart glasses, and ground-level traffic lights or signs were discussed.}, journal={APPLIED ERGONOMICS}, publisher={Elsevier BV}, author={Yuan, Jing and Bai, Xiaolu and Driscoll, Brendan and Liu, Ming and Huang, He and Feng, Jing}, year={2022}, month={Oct} }
 @article{lee_liu_lewek_hu_filer_huang_2022, title={Toward Safe Wearer-Prosthesis Interaction: Evaluation of Gait Stability and Human Compensation Strategy Under Faults in Robotic Transfemoral Prostheses}, volume={30}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85139401676&partnerID=MN8TOARS}, DOI={10.1109/TNSRE.2022.3208778}, abstractNote={Although advanced wearable robots can assist human wearers, their internal faults (i.e., sensors or control errors) also pose a challenge. To ensure safe wearer-robot interactions, how internal errors by the prosthesis limb affect the stability of the user-prosthesis system, and how users react and compensate for the instability elicited by internal errors are imperative. The goals of this study were to 1) systematically investigate the biomechanics of a wearer-robot system reacting to internal errors induced by a powered knee prosthesis (PKP), and 2) quantify the error tolerable bound that does not affect the user’s gait stability. Eight non-disabled participants and two unilateral transfemoral amputees walked on a pathway wearing a PKP, as the controller randomly switched the control parameters to disturbance parameters to mimic the errors caused by locomotion mode misrecognition. The size of prosthesis control errors was systematically varied to determine the error tolerable bound that disrupted gait stability. The effect of the error was quantified based on the 1) mechanical change described by the angular impulse applied by the PKP, and 2) overall gait instability quantified using human perception, angular momentum, and compensatory stepping. The results showed that the error tolerable bound is dependent on the gait phase and the direction of torque change. Two balance recovery strategies were also observed to allow participants to successful respond to the induced errors. The outcomes of this study may assist the future design of an auto-tuning algorithm, volitionally-controlled powered prosthetic legs, and training of gait stability.}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Lee, I-Chieh and Liu, Ming and Lewek, Michael D. and Hu, Xiaogang and Filer, William G. and Huang, He}, year={2022}, pages={2773–2782} }
 @article{yao_zhou_hinson_dong_wu_ives_hu_huang_zhu_2022, title={Ultrasoft Porous 3D Conductive Dry Electrodes for Electrophysiological Sensing and Myoelectric Control}, volume={5}, ISSN={["2365-709X"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85132598682&partnerID=MN8TOARS}, DOI={10.1002/admt.202101637}, abstractNote={AbstractBiopotential electrodes have found broad applications in health monitoring, human–machine interactions, and rehabilitation. This article reports the fabrication and applications of ultrasoft breathable dry electrodes that can address several challenges for their long‐term wearable applications—skin compatibility, wearability, and long‐term stability. The proposed electrodes rely on porous and conductive silver nanowire‐based nanocomposites as the robust mechanical and electrical interface. The highly conductive and conformable structure eliminates the necessity of conductive gel while establishing a sufficiently low electrode–skin impedance for high‐fidelity electrophysiological sensing. The introduction of gas‐permeable structures via a simple and scalable method based on sacrificial templates improves breathability and skin compatibility for applications requiring long‐term skin contact. Such conformable and breathable dry electrodes allow for efficient and unobtrusive monitoring of heart, muscle, and brain activities. In addition, based on the muscle activities captured by the electrodes and a musculoskeletal model, electromyogram‐based neural–machine interfaces are realized, illustrating the great potential for prosthesis control, neurorehabilitation, and virtual reality.}, number={10}, journal={ADVANCED MATERIALS TECHNOLOGIES}, author={Yao, Shanshan and Zhou, Weixin and Hinson, Robert and Dong, Penghao and Wu, Shuang and Ives, Jasmine and Hu, Xiaogang and Huang, He and Zhu, Yong}, year={2022}, month={May} }
 @article{tu_li_liu_si_huang_2021, title={A Data-Driven Reinforcement Learning Solution Framework for Optimal and Adaptive Personalization of a Hip Exoskeleton}, volume={2021-May}, ISSN={["2577-087X"]}, url={http://dx.doi.org/10.1109/icra48506.2021.9562062}, DOI={10.1109/icra48506.2021.9562062}, abstractNote={Robotic exoskeletons are exciting technologies for augmenting human mobility. However, designing such a device for seamless integration with the human user and to assist human movement still is a major challenge. This paper aims at developing a novel data-driven solution framework based on reinforcement learning (RL), without first modeling the human-robot dynamics, to provide optimal and adaptive personalized torque assistance for reducing human efforts during walking. Our automatic personalization solution framework includes the assistive torque profile with two control timing parameters (peak and offset timings), the least square policy iteration (LSPI) for learning the parameter tuning policy, and a cost function based on a transferred work ratio. The proposed controller was successfully validated on a healthy human subject to assist unilateral hip extension in walking. The results showed that the optimal and adaptive RL controller as a new approach was feasible for tuning assistive torque profile of the hip exoskeleton that coordinated with human actions and reduced activation level of hip extensor muscle in human.}, journal={2021 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA 2021)}, publisher={IEEE}, author={Tu, Xikai and Li, Minhan and Liu, Ming and Si, Jennie and Huang, He}, year={2021}, pages={10610–10616} }
 @article{upadhye_shah_liu_buckner_huang_2021, title={A Powered Prosthetic Ankle Designed for Task Variability - A Concept Validation}, ISSN={["2153-0858"]}, url={http://dx.doi.org/10.1109/iros51168.2021.9636324}, DOI={10.1109/iros51168.2021.9636324}, abstractNote={Ankle joints play key roles in everyday locomotion, such as walking, stair climbing, and sit-to-stand. Despite the achievement in designing powered prosthetic ankles, engineers still face challenges to duplicate the full mechanics of ankle joints, including high torque, large range of motion (ROM), low profile, backdrivability, and efficiency, using electric motors and related transmissions. In this study, our goal was to develop a new active prosthetic ankle, Variable Spring embedded Motor-ball screw (VSeM) ankle, to meet all these requirements at the same time. Using a manually adjustable elastic element, which is parallel with our motor actuator, we can readjust the ROM of VSeM to handle all normal locomotion tasks. VSeM’s capability to mimic human ankle was validated through both bench tests and human subject tests.}, journal={2021 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS)}, publisher={IEEE}, author={Upadhye, Sameer and Shah, Chinmay and Liu, Ming and Buckner, Gregory and Huang, He}, year={2021}, pages={6153–6158} }
 @article{pan_huang_2021, title={A robust model-based neural-machine interface across different loading weights applied at distal forearm}, volume={67}, ISSN={["1746-8108"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85101330574&partnerID=MN8TOARS}, DOI={10.1016/j.bspc.2021.102509}, abstractNote={Musculoskeletal models (MMs) have recently been proposed to decode electromyography (EMG) signals for movement intent recognition. Since the robustness is critical to retain the performance of neural-machine interface (NMI) during daily activities and the loading weight change is one of the critical factors that would affect the performance of NMI, this study aimed to further investigate the robustness of a generic MM-based NMI across different loading conditions. Eight able-bodied (AB) individuals and one individual with a transradial amputation were recruited and tested while performing a real-time virtual wrist/hand posture matching task under different loading weights (AB subjects: 0 kg, 0.567 kg, and 1.134 kg; amputee subject: 0 kg and 0.567 kg) applied at the distal forearm. All tasks were achieved by both AB individuals and the individual with the transradial amputation. There was no significant difference among the real-time performance (completion time, the number of overshoots, and path efficiency) of AB individuals under different loading conditions. We calculated the average muscle activations of each muscle during the initial 0.5 s and last 0.5 s respectively for each target across all subjects and trials. The analysis of muscle activations showed that additional weights caused muscle co-contractions. However, the subjects can cope with the increased muscle co-activation level, modifying muscle activation patterns, and still complete tasks successfully. We obtained similar results from the individual with the transradial amputation. These results demonstrated the robustness of MM-based NMI across different loading conditions. The outcomes indicate the potential of the multi-user NMI toward practical applications.}, journal={BIOMEDICAL SIGNAL PROCESSING AND CONTROL}, author={Pan, Lizhi and Huang, He}, year={2021}, month={May} }
 @article{vargas_huang_zhu_hu_2021, title={Closed-loop control of a prosthetic finger via evoked proprioceptive information}, volume={18}, ISSN={["1741-2552"]}, url={http://dx.doi.org/10.1088/1741-2552/ac3c9e}, DOI={10.1088/1741-2552/ac3c9e}, abstractNote={Abstract
               
                  Objective. Proprioceptive information plays an important role for recognizing and coordinating our limb’s static and dynamic states relative to our body or the environment. In this study, we determined how artificially evoked proprioceptive feedback affected the continuous control of a prosthetic finger. Approach. We elicited proprioceptive information regarding the joint static position and dynamic movement of a prosthetic finger via a vibrotactor array placed around the subject’s upper arm. Myoelectric signals of the finger flexor and extensor muscles were used to control the prosthesis, with or without the evoked proprioceptive feedback. Two control modes were evaluated: the myoelectric signal amplitudes were continuously mapped to either the position or the velocity of the prosthetic joint. Main results. Our results showed that the evoked proprioceptive information improved the control accuracy of the joint angle, with comparable performance in the position- and velocity-control conditions. However, greater angle variability was prominent during position-control than velocity-control. Without the proprioceptive feedback, the position-control tended to show a smaller angle error than the velocity-control condition. Significance. Our findings suggest that closed-loop control of a prosthetic device can potentially be achieved using non-invasive evoked proprioceptive feedback delivered to intact participants. Moreover, the evoked sensory information was integrated during myoelectric control effectively for both control strategies. The outcomes can facilitate our understanding of the sensorimotor integration process during human-machine interactions, which can potentially promote fine control of prosthetic hands.}, number={6}, journal={JOURNAL OF NEURAL ENGINEERING}, publisher={IOP Publishing}, author={Vargas, Luis and Huang, He and Zhu, Yong and Hu, Xiaogang}, year={2021}, month={Dec} }
 @misc{yuan_cline_liu_huang_feng_2021, title={Cognitive measures during walking with and without lower-limb prosthesis: protocol for a scoping review}, volume={11}, ISBN={2044-6055}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85101189834&partnerID=MN8TOARS}, DOI={10.1136/bmjopen-2020-039975}, abstractNote={IntroductionTuning of lower-limb (LL) robotic prosthesis control is necessary to provide personalised assistance to each human wearer during walking. Prostheses wearers’ adaptation processes are subjective and the efficiency largely depends on one’s mental processes. Therefore, beyond physical motor performance, prosthesis personalisation should consider the wearer’s preference and cognitive performance during walking. As a first step, it is necessary to examine the current measures of cognitive performance when a wearer walks with an LL prosthesis, identify the gaps and methodological considerations, and explore additional measures in a walking setting. In this protocol, we outlined a scoping review that will systematically summarise and evaluate the measures of cognitive performance during walking with and without LL prosthesis.Methods and analysisThe review process will be guided and documented by CADIMA, an open-access online data management portal for evidence synthesis. Keyword searches will be conducted in seven databases (Web of Science, MEDLINE, BIOSIS, SciELO Citation Index, ProQuest, CINAHL and PsycINFO) up to 2020 supplemented with grey literature searches. Retrieved records will be screened by at least two independent reviewers on the title-and-abstract level and then the full-text level. Selected studies will be evaluated for reporting bias. Data on sample characteristics, type of cognitive function, characteristics of cognitive measures, task prioritisation, experimental design and walking setting will be extracted.Ethics and disseminationThis scoping review will evaluate the measures used in previously published studies thus does not require ethical approval. The results will contribute to the advancement of prosthesis tuning processes by reviewing the application status of cognitive measures during walking with and without prosthesis and laying the foundation for developing needed measures for cognitive assessment during walking. The results will be disseminated through conferences and journals.}, number={2}, journal={BMJ OPEN}, author={Yuan, Jing and Cline, Emily and Liu, Ming and Huang, He and Feng, Jing}, year={2021} }
 @article{rubin_liu_hu_huang_2021, title={Common Neural Input within and across Lower Limb Muscles: A Preliminary Study}, ISSN={["1558-4615"]}, url={http://dx.doi.org/10.1109/embc46164.2021.9630141}, DOI={10.1109/embc46164.2021.9630141}, abstractNote={Motor units (MUs) are the basic unit of motor control. MU synchronization has been evaluated to identify common inputs in neural circuitry during motor coordination. Recent studies have compared common inputs between muscles in the lower limb, but further investigation is needed to compare common inputs to MUs both within a muscle and between MUs of different muscle pairs. The goal of this preliminary study was to characterize levels of common inputs to MUs in three muscle groups: MUs within a muscle, between bilateral homologous pairs, and between agonist/antagonist muscle pairs. To achieve this, surface electromyography (EMG) was recorded during bilateral ankle dorsiflexion and plantarflexion on the right and left tibiales anterior (RTA, LTA) and gastrocnemii (RGA, LGA) muscles. After decomposing EMG into active MU firings, we conducted coherence analyses of composite MU spike trains (CSTs) in each muscle group in both the beta (13-30 Hz) and gamma (30-60 Hz) frequency bands. Our results indicate MUs within a muscle have the greatest levels of common input, with decreasing levels of common input to bilateral and agonist/antagonist muscle pairs, respectively. Additionally, each muscle group exhibited similar levels of common input between the beta and gamma bands. This work may provide a way to unveil mechanisms of functional coordination in the lower limb across motor tasks.}, journal={2021 43RD ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE & BIOLOGY SOCIETY (EMBC)}, publisher={IEEE}, author={Rubin, Noah and Liu, Wentao and Hu, Xiaogang and Huang, He}, year={2021}, pages={6683–6686} }
 @article{berman_hinson_huang_2021, title={Comparing Reinforcement Learning Agents and Supervised Learning Neural Networks for EMG-Based Decoding of Continuous Movements}, ISSN={["1558-4615"]}, url={http://dx.doi.org/10.1109/embc46164.2021.9630744}, DOI={10.1109/embc46164.2021.9630744}, abstractNote={Recent work on electromyography (EMG)-based decoding of continuous joint kinematics has included model-based approaches, such as musculoskeletal modeling, as well as model-free approaches such as supervised learning neural networks (SLNN). This study aimed to present a new kinematics decoding framework based on reinforcement learning (RL), which combines machine learning and model-based approaches together. We compared the performance and robustness of our new method with those of the SLNN approach. EMG and kinematic data were collected from 5 able-bodied subjects while they performed flexion and extension of the metacarpophalangeal (MCP) and wrist joints simultaneously at both a slow and fast tempo. The data were used to train an RL agent and a SLNN for each of the 2 tempos. All the trained agents and SLNNs were tested with both fast and slow kinematic data. Pearson’s correlation coefficient (r) and normalized root mean square error (NRMSE) between measured and estimated joint angles were used to determine performance. Our results suggest that the RL-based kinematics decoder is more robust to changes in movement speeds between training and testing data and has better performance than the SLNN.}, journal={2021 43RD ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE & BIOLOGY SOCIETY (EMBC)}, publisher={IEEE}, author={Berman, Joseph and Hinson, Robert and Huang, He}, year={2021}, pages={6297–6300} }
 @article{popp_liu_huang_2021, title={Development of a Wearable Human-Machine Interface to Track Forearm Rotation via an Optical Sensor}, ISSN={["1558-4615"]}, url={http://dx.doi.org/10.1109/embc46164.2021.9629851}, DOI={10.1109/embc46164.2021.9629851}, abstractNote={The goal of this research was to develop an intuitive wearable human-machine interface (HMI), utilizing an optical sensor. The proposed system quantifies wrist pronation and supination using an optical displacement sensor. Compared with existing systems, this HMI ensures intuitiveness by relying on direct measurement of forearm position, minimizes involved sensors, and is expected to be long-lasting. To test for feasibility, the developed HMI was implemented to control a prosthetic wrist based on forearm rotation of able-bodied subjects. Performance of optical sensor system (OSS) prosthesis control was compared to electromyography (EMG) based direct control, for six able-bodied individuals, using a clothespin relocation task. Results showed that the performance of OSS control was comparable to direct control, therefore validating the feasibility of the OSS HMI.}, journal={2021 43RD ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE & BIOLOGY SOCIETY (EMBC)}, publisher={IEEE}, author={Popp, Fiona and Liu, Ming and Huang, He}, year={2021}, pages={7360–7363} }
 @article{liu_fleming_lee_huang_2021, title={Direct Myoelectric Control Modifies Lower Limb Functional Connectivity: A Case Study}, ISSN={["1558-4615"]}, url={http://dx.doi.org/10.1109/embc46164.2021.9630844}, DOI={10.1109/embc46164.2021.9630844}, abstractNote={Prostheses with direct EMG control could restore amputee’s biomechanics structure and residual muscle functions by using efferent signals to drive prosthetic ankle joint movements. Because only feedforward control is restored, it is unclear 1) what neuromuscular control mechanisms are used in coordinating residual and intact muscle activities and 2) how this mechanism changes over guided training with the prosthetic ankle. To address these questions, we applied functional connectivity analysis to an individual with unilateral lower-limb amputation during postural sway task. We built functional connectivity networks of surface EMGs from eleven lower-limb muscles during three sessions to investigate the coupling among different function modules. We observed that functional network was reshaped by training and we identified a stronger connection between residual and intact below knee modules with improved bilateral symmetry after amputee acquired skills to better control the powered prosthetic ankle. The evaluation session showed that functional connectivity was largely preserved even after nine months interval. This preliminary study might inform a unique way to unveil the potential neuromechanic changes that occur after extended training with direct EMG control of a powered prosthetic ankle.}, journal={2021 43RD ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE & BIOLOGY SOCIETY (EMBC)}, publisher={IEEE}, author={Liu, Wentao and Fleming, Aaron and Lee, I-Chieh and Huang, He Helen}, year={2021}, pages={6573–6576} }
 @article{fleming_huang_buxton_hodges_huang_2021, title={Direct continuous electromyographic control of a powered prosthetic ankle for improved postural control after guided physical training: A case study}, volume={2}, url={http://dx.doi.org/10.1017/wtc.2021.2}, DOI={10.1017/wtc.2021.2}, abstractNote={Abstract
	  Despite the promise of powered lower limb prostheses, existing controllers do not assist many daily activities that require continuous control of prosthetic joints according to human states and environments. The objective of this case study was to investigate the feasibility of direct, continuous electromyographic (dEMG) control of a powered ankle prosthesis, combined with physical therapist-guided training, for improved standing postural control in an individual with transtibial amputation. Specifically, EMG signals of the residual antagonistic muscles (i.e. lateral gastrocnemius and tibialis anterior) were used to proportionally drive pneumatical artificial muscles to move a prosthetic ankle. Clinical-based activities were used in the training and evaluation protocol of the control paradigm. We quantified the EMG signals in the bilateral shank muscles as well as measures of postural control and stability. Compared to the participant’s daily passive prosthesis, the dEMG-controlled ankle, combined with the training, yielded improved clinical balance scores and reduced compensation from intact joints. Cross-correlation coefficient of bilateral center of pressure excursions, a metric for quantifying standing postural control, increased to .83(±.07) when using dEMG ankle control (passive device: .39(±.29)). We observed synchronized activation of homologous muscles, rapid improvement in performance on the first day of the training for load transfer tasks, and further improvement in performance across training days (p = .006). This case study showed the feasibility of this dEMG control paradigm of a powered prosthetic ankle to assist postural control. This study lays the foundation for future study to extend these results through the inclusion of more participants and activities.}, journal={Wearable Technologies}, publisher={Cambridge University Press (CUP)}, author={Fleming, Aaron and Huang, Stephanie and Buxton, Elizabeth and Hodges, Frank and Huang, He Helen}, year={2021} }
 @article{nalam_huang_2021, title={Empowering prosthesis users with a hip exoskeleton}, volume={27}, ISSN={["1546-170X"]}, url={http://dx.doi.org/10.1038/s41591-021-01529-w}, DOI={10.1038/s41591-021-01529-w}, number={10}, journal={NATURE MEDICINE}, publisher={Springer Science and Business Media LLC}, author={Nalam, Varun and Huang, He}, year={2021}, month={Oct}, pages={1677–1678} }
 @article{zhong_silva_li_huang_lobaton_2021, title={Environmental Context Prediction for Lower Limb Prostheses With Uncertainty Quantification}, volume={18}, ISSN={["1558-3783"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85097863425&partnerID=MN8TOARS}, DOI={10.1109/TASE.2020.2993399}, abstractNote={Reliable environmental context prediction is critical for wearable robots (e.g., prostheses and exoskeletons) to assist terrain-adaptive locomotion. This article proposed a novel vision-based context prediction framework for lower limb prostheses to simultaneously predict human’s environmental context for multiple forecast windows. By leveraging the Bayesian neural networks (BNNs), our framework can quantify the uncertainty caused by different factors (e.g., observation noise, and insufficient or biased training) and produce a calibrated predicted probability for online decision-making. We compared two wearable camera locations (a pair of glasses and a lower limb device), independently and conjointly. We utilized the calibrated predicted probability for online decision-making and fusion. We demonstrated how to interpret deep neural networks with uncertainty measures and how to improve the algorithms based on the uncertainty analysis. The inference time of our framework on a portable embedded system was less than 80 ms/frame. The results in this study may lead to novel context recognition strategies in reliable decision-making, efficient sensor fusion, and improved intelligent system design in various applications. Note to Practitioners—This article was motivated by two practical problems in computer vision for wearable robots: First, the performance of deep neural networks is challenged by real-life disturbances. However, reliable confidence estimation is usually unavailable and the factors causing failures are hard to identify. Second, evaluating wearable robots by intuitive trial and error is expensive due to the need for human experiments. Our framework produces a calibrated predicted probability as well as three uncertainty measures. The calibrated probability makes it easy to customize prediction decision criteria by considering how much the corresponding application can tolerate error. This study demonstrated a practical procedure to interpret and improve the performance of deep neural networks with uncertainty quantification. We anticipate that our methodology could be extended to other applications as a general scientific and efficient procedure of evaluating and improving intelligent systems.}, number={2}, journal={IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Zhong, Boxuan and Silva, Rafael Luiz and Li, Minhan and Huang, He and Lobaton, Edgar}, year={2021}, month={Apr}, pages={458–470} }
 @article{lee_fylstra_liu_lenzi_huang_2021, title={Is there a Trade-off between Economy and Task Goal Variability in Transfemoral Amputee Gait?}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85132398213&partnerID=MN8TOARS}, DOI={10.21203/rs.3.rs-659520}, journal={ResearchSquare}, author={Lee, I.C. and Fylstra, B.L. and Liu, M. and Lenzi, T. and Huang, H.}, year={2021} }
 @misc{fleming_stafford_huang_hu_ferris_huang_2021, title={Myoelectric control of robotic lower limb prostheses: a review of electromyography interfaces, control paradigms, challenges and future directions}, volume={18}, ISSN={["1741-2552"]}, url={http://dx.doi.org/10.1088/1741-2552/ac1176}, DOI={10.1088/1741-2552/ac1176}, abstractNote={Abstract
               
                  Objective. Advanced robotic lower limb prostheses are mainly controlled autonomously. Although the existing control can assist cyclic movements during locomotion of amputee users, the function of these modern devices is still limited due to the lack of neuromuscular control (i.e. control based on human efferent neural signals from the central nervous system to peripheral muscles for movement production). Neuromuscular control signals can be recorded from muscles, called electromyographic (EMG) or myoelectric signals. In fact, using EMG signals for robotic lower limb prostheses control has been an emerging research topic in the field for the past decade to address novel prosthesis functionality and adaptability to different environments and task contexts. The objective of this paper is to review robotic lower limb Prosthesis control via EMG signals recorded from residual muscles in individuals with lower limb amputations. Approach. We performed a literature review on surgical techniques for enhanced EMG interfaces, EMG sensors, decoding algorithms, and control paradigms for robotic lower limb prostheses. Main results. This review highlights the promise of EMG control for enabling new functionalities in robotic lower limb prostheses, as well as the existing challenges, knowledge gaps, and opportunities on this research topic from human motor control and clinical practice perspectives. Significance. This review may guide the future collaborations among researchers in neuromechanics, neural engineering, assistive technologies, and amputee clinics in order to build and translate true bionic lower limbs to individuals with lower limb amputations for improved motor function.}, number={4}, journal={JOURNAL OF NEURAL ENGINEERING}, publisher={IOP Publishing}, author={Fleming, Aaron and Stafford, Nicole and Huang, Stephanie and Hu, Xiaogang and Ferris, Daniel P. and Huang, He}, year={2021}, month={Aug} }
 @inbook{fleming_liu_huang_2022, title={Neural Coherence of Homologous Muscle Pairs During Direct EMG Control of Standing Posture in Transtibial Amputees}, volume={28}, url={http://dx.doi.org/10.1007/978-3-030-70316-5_23}, DOI={10.1007/978-3-030-70316-5_23}, abstractNote={The objective of this preliminary study was to investigate the feasibility of a transtibial amputee to improve interlimb muscle coordination while using direct, continuous control of a powered ankle prosthesis combined with physical therapist guided training, for improved standing postural control. A participant with transtibial amputation received an extended PT-guided training on posture while using the dEMG control of powered ankle with his residual Tibialis Anterior and Lateral Gastrocnemius. We quantified cross-correlation of Center of Pressure excursions and coherence in EMG signals from the bilateral shank muscles. Between-limb coordination was observed as synchronized activation of homologous muscles in the shank. We observed increased coherence in the TA muscle pair after training in 0–5 Hz and 10–20 Hz frequency ranges. These results demonstrate the potential for amputees to closely coordinate residual muscle activations with intact muscles given sufficient training. It is further possible amputees adapt sources of descending neural commands between homologous muscle pairs after guided training. Future study requires more participants to validate these results.}, booktitle={Biosystems & Biorobotics}, publisher={Springer International Publishing}, author={Fleming, Aaron and Liu, Wentao and Huang, He}, year={2022}, pages={139–143} }
 @article{vargas_huang_zhu_hu_2022, title={Object Recognition via Evoked Sensory Feedback during Control of a Prosthetic Hand}, volume={7}, ISSN={["2377-3766"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85118588970&partnerID=MN8TOARS}, DOI={10.1109/LRA.2021.3122897}, abstractNote={Haptic and proprioceptive feedback is critical for sensorimotor integration when we use our hand to perform daily tasks. Here, we evaluated how externally evoked haptic and proprioceptive feedback and myoelectric control strategies affected the recognition of object properties when participants controlled a prosthetic hand. Fingertip haptic sensation was elicited using a transcutaneous nerve stimulation grid to encode the prosthetic's fingertip forces. An array of tactors elicited patterned vibratory stimuli to encode tactile-proprioceptive kinematic information of the prosthetic finger joint. Myoelectric signals of the finger flexor and extensor were used to control the position or velocity of joint angles of the prosthesis. Participants were asked to perform object property (stiffness and size) recognition, by controlling the prosthetic hand with concurrent haptic and tactile-proprioceptive feedback. With the evoked feedback, intact and amputee participants recognized the object stiffness and size at success rates ranging from 50% to 100% in both position and velocity control with no significant difference across control schemes. Our findings show that evoked somatosensory feedback in a non-invasive manner can facilitate closed-loop control of the prosthetic hand and allowed for simultaneous recognition of different object properties. The outcomes can facilitate our understanding on the role of sensory feedback during bidirectional human-machine interactions, which can potentially promote user experience in object interactions using prosthetic hands.}, number={1}, journal={IEEE ROBOTICS AND AUTOMATION LETTERS}, author={Vargas, Luis and Huang, He and Zhu, Yong and Hu, Xiaogang}, year={2022}, month={Jan}, pages={207–214} }
 @article{vargas_huang_zhu_hu_2021, title={Perception of Static Position and Kinesthesia of the Finger using Vibratory Stimulation}, volume={2021-May}, ISSN={["1948-3546"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85107464376&partnerID=MN8TOARS}, DOI={10.1109/NER49283.2021.9441255}, abstractNote={Proprioception provides information regarding the state of an individual's limb in terms of static position and kinesthesia (dynamic movement). When such feedback is lost or impaired, the performance of dexterous control of our biological limbs or assistive devices tends to deteriorate. In this study, we determined if external vibratory stimulation patterns could allow for the perception of a finger's static position and kinesthesia. Using four tactors and two stimulus levels, eight vibratory settings corresponded to eight discrete finger positions. The transition patterns between these eight settings corresponded to kinesthesia. Three experimental blocks assessed the perception of a finger's static position, speed, and movement (amplitude and direction). Our results demonstrated that both position and kinesthesia could be recognized with over 93% accuracy. The outcomes suggest that vibratory stimulus can inform subjects of static and dynamic aspects of finger proprioception. This sensory stimulation approach can be implemented to improve outcomes in clinical populations with sensory deficits, and to enhance user experience when users interact with assistive devices.}, journal={2021 10TH INTERNATIONAL IEEE/EMBS CONFERENCE ON NEURAL ENGINEERING (NER)}, author={Vargas, Luis and Huang, He and Zhu, Yong and Hu, Xiaogang}, year={2021}, pages={1087–1090} }
 @article{wu_li_yao_si_huang_2021, title={Reinforcement learning enabled automatic impedance control of a robotic knee prosthesis to mimic the intact knee motion in a co-adapting environment}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85101254246&partnerID=MN8TOARS}, journal={arXiv}, author={Wu, R. and Li, M. and Yao, Z. and Si, J. and Huang, H.}, year={2021} }
 @article{huang_wu_yao_si_2021, title={Robotic knee tracking control to mimic the intact human knee profile based on actor-critic reinforcement learning}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85101456660&partnerID=MN8TOARS}, journal={arXiv}, author={Huang, H. and Wu, R. and Yao, Z. and Si, J.}, year={2021} }
 @article{vargas_huang_zhu_hu_2021, title={Static and dynamic proprioceptive recognition through vibrotactile stimulation}, volume={18}, ISSN={["1741-2552"]}, url={http://dx.doi.org/10.1088/1741-2552/ac0d43}, DOI={10.1088/1741-2552/ac0d43}, abstractNote={Objective. Proprioceptive information provides individuals with a sense of our limb’s static position and dynamic movement. Impaired or a lack of such feedback can diminish our ability to perform dexterous motions with our biological limbs or assistive devices. Here we seek to determine whether both static and dynamic components of proprioception can be recognized using variation of the spatial and temporal components of vibrotactile feedback. Approach. An array of five vibrotactors was placed on the forearm of each subject. Each tactor was encoded to represent one of the five forearm postures. Vibratory stimulus was elicited to convey the static position and movement of the forearm. Four experimental blocks were performed to test each subject’s recognition of a forearm’s simulated static position, rotational amplitude, rotational amplitude and direction, and rotational speed. Main results. Our results showed that the subjects were able to perform proprioceptive recognition based on the delivered vibrotactile information. Specifically, rotational amplitude recognition resulted in the highest level of accuracy (99.0%), while the recognition accuracy of the static position and the rotational amplitude-direction was the lowest (91.7% and 90.8%, respectively). Nevertheless, all proprioceptive properties were perceived with >90% accuracy, indicating that the implemented vibrotactile encoding scheme could effectively provide proprioceptive information to the users. Significance. The outcomes suggest that information pertaining to static and dynamic aspects of proprioception can be accurately delivered using an array of vibrotactors. This feedback approach could be used to potentially evaluate the sensorimotor integration processes during human–machine interactions, and to improve sensory feedback in clinical populations with somatosensory impairments.}, number={4}, journal={JOURNAL OF NEURAL ENGINEERING}, publisher={IOP Publishing}, author={Vargas, Luis and Huang, He and Zhu, Yong and Hu, Xiaogang}, year={2021}, month={Aug} }
 @article{huang_si_brandt_li_2021, title={Taking both sides: seeking symbiosis between intelligent prostheses and human motor control during locomotion}, volume={20}, ISSN={["2468-4511"]}, url={http://dx.doi.org/10.1016/j.cobme.2021.100314}, DOI={10.1016/j.cobme.2021.100314}, abstractNote={Robotic lower-limb prostheses aim to replicate the power-generating capability of biological joints during locomotion to empower individuals with lower-limb loss. However, recent clinical trials have not demonstrated clear advantages of these devices over traditional passive devices. We believe this is partly because the current designs of robotic prothesis controllers and clinical methods for fitting and training individuals to use them do not ensure good coordination between the prosthesis and user. Accordingly, we advocate for new holistic approaches in which human motor control and intelligent prosthesis control function as one system (defined as human-prosthesis symbiosis). We hope engineers and clinicians will work closely to achieve this symbiosis, thereby improving the functionality and acceptance of robotic prostheses and users' quality of life.}, journal={CURRENT OPINION IN BIOMEDICAL ENGINEERING}, publisher={Elsevier BV}, author={Huang, He and Si, Jennie and Brandt, Andrea and Li, Minhan}, year={2021}, month={Dec} }
 @article{tabor_agcayazi_fleming_thompson_kapoor_liu_lee_huang_bozkurt_ghosh_2021, title={Textile-Based Pressure Sensors for Monitoring Prosthetic-Socket Interfaces}, volume={21}, ISSN={["1558-1748"]}, url={http://dx.doi.org/10.1109/jsen.2021.3053434}, DOI={10.1109/JSEN.2021.3053434}, abstractNote={Amputees are prone to experiencing discomfort when wearing their prosthetic devices. As the amputee population grows this becomes a more prevalent and pressing concern. There is a need for new prosthetic technologies to construct more comfortable and well-fitted liners and sockets. One of the well-recognized impediments to the development of new prosthetic technology is the lack of practical inner socket sensors to monitor the inner socket environment (ISE), or the region between the residual limb and the socket. Here we present a capacitive pressure sensor fabricated through a simple, and scalable sewing process using commercially available conductive yarns and textile materials. This fully-textile sensor provides a soft, flexible, and comfortable sensing system for monitoring the ISE. We provide details of our low-power sensor system capable of high-speed data collection from up to four sensor arrays. Additionally, we demonstrate two custom set-ups to test and validate the textile-based sensors in a simulated prosthetic environment. Finally, we utilize the textile-based sensors to study the ISE of a bilateral transtibial amputee. Results indicate that the textile-based sensors provide a promising potential for seamlessly monitoring the ISE.}, number={7}, journal={IEEE SENSORS JOURNAL}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Tabor, Jordan and Agcayazi, Talha and Fleming, Aaron and Thompson, Brendan and Kapoor, Ashish and Liu, Ming and Lee, Michael Y. and Huang, He and Bozkurt, Alper and Ghosh, Tushar K.}, year={2021}, month={Apr}, pages={9413–9422} }
 @article{li_wen_gao_si_huang_2021, title={Toward Expedited Impedance Tuning of a Robotic Prosthesis for Personalized Gait Assistance by Reinforcement Learning Control}, volume={38}, ISSN={["1941-0468"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85107219950&partnerID=MN8TOARS}, DOI={10.1109/TRO.2021.3078317}, abstractNote={Personalizing medical devices such as lower limb wearable robots is challenging. While the initial feasibility of automating the process of knee prosthesis control parameter tuning has been demonstrated in a principled way, the next critical issue is to improve tuning efficiency and speed it up for the human user, in clinic settings, while maintaining human safety. We, therefore, propose a policy iteration with constraint embedded (PICE) method as an innovative solution to the problem under the framework of reinforcement learning. Central to PICE is the use of a projected Bellman equation with a constraint of assuring positive semidefiniteness of performance values during policy evaluation. Additionally, we developed both online and offline PICE implementations that provide additional flexibility for the designer to fully utilize measurement data, either from on-policy or off-policy, to further improve PICE tuning efficiency. Our human subject testing showed that the PICE provided effective policies with significantly reduced tuning time. For the first time, we also experimentally evaluated and demonstrated the robustness of the deployed policies by applying them to different tasks and users. Putting it together, our new way of problem solving has been effective as PICE has demonstrated its potential toward truly automating the process of control parameter tuning for robotic knee prosthesis users.}, number={1}, journal={IEEE TRANSACTIONS ON ROBOTICS}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Li, Minhan and Wen, Yue and Gao, Xiang and Si, Jennie and Huang, He}, year={2021}, month={May} }
 @article{farina_vujaklija_branemark_bull_dietl_graimann_hargrove_hoffmann_huang_ingvarsson_et al._2021, title={Toward higher-performance bionic limbs for wider clinical use}, volume={5}, ISSN={["2157-846X"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85107532112&partnerID=MN8TOARS}, DOI={10.1038/s41551-021-00732-x}, abstractNote={Most prosthetic limbs can autonomously move with dexterity, yet they are not perceived by the user as belonging to their own body. Robotic limbs can convey information about the environment with higher precision than biological limbs, but their actual performance is substantially limited by current technologies for the interfacing of the robotic devices with the body and for transferring motor and sensory information bidirectionally between the prosthesis and the user. In this Perspective, we argue that direct skeletal attachment of bionic devices via osseointegration, the amplification of neural signals by targeted muscle innervation, improved prosthesis control via implanted muscle sensors and advanced algorithms, and the provision of sensory feedback by means of electrodes implanted in peripheral nerves, should all be leveraged towards the creation of a new generation of high-performance bionic limbs. These technologies have been clinically tested in humans, and alongside mechanical redesigns and adequate rehabilitation training should facilitate the wider clinical use of bionic limbs.}, journal={NATURE BIOMEDICAL ENGINEERING}, author={Farina, Dario and Vujaklija, Ivan and Branemark, Rickard and Bull, Anthony M. J. and Dietl, Hans and Graimann, Bernhard and Hargrove, Levi J. and Hoffmann, Klaus-Peter and Huang, He and Ingvarsson, Thorvaldur and et al.}, year={2021}, month={May} }
 @inbook{wu_li_si_huang_2022, title={Understanding Human-Prosthesis Interaction via Reinforcement Learning-Based Echo Control: A Case Study}, volume={28}, url={http://dx.doi.org/10.1007/978-3-030-70316-5_112}, DOI={10.1007/978-3-030-70316-5_112}, abstractNote={This case study aimed to understand human-prosthesis interaction while the impedance control of a robotic prosthesis was tuned in order to echo the knee kinematics on the intact joint. Echo control derives from a common belief that if the prosthesis joint mechanics meet those of the intact joint, more symmetrical and normal gait should be reached in the prosthesis user. In this study, our previous developed reinforcement learning (RL) control was used to tune impedance of a power knee prosthesis in walking to achieve echo control. It was tested on one able-bodied human subject walking with the robotic knee. The results showed that the prosthesis control parameter tuning was coupled with changes in intact knee mechanics. Nevertheless, regardless such neuromechanic coupling between the two lower limbs, RL was robust to tune prosthesis control and meet the intact knee kinematics. Finally, the RL echo control enabled us to examine gait symmetry. Additional research efforts are still needed to identify the influence of echo control of prosthetic knee on gait tempospatial symmetry.}, booktitle={Biosystems & Biorobotics}, publisher={Springer International Publishing}, author={Wu, Ruofan and Li, Minhan and Si, Jennie and Huang, He}, year={2022}, pages={697–701} }
 @article{alili_nalam_li_liu_si_huang_2021, title={User Controlled Interface for Tuning Robotic Knee Prosthesis}, ISSN={["2153-0858"]}, url={http://dx.doi.org/10.1109/iros51168.2021.9636264}, DOI={10.1109/iros51168.2021.9636264}, abstractNote={The tuning process for a robotic prosthesis is a challenging and time-consuming task both for users and clinicians. An automatic tuning approach using reinforcement learning (RL) has been developed for a knee prosthesis to address the challenges of manual tuning methods. The algorithm tunes the optimal control parameters based on the provided knee joint profile that the prosthesis is expected to replicate during gait safely. This paper presents an intuitive interface designed for the prosthesis users and clinicians to choose the preferred knee joint profile during gait and use the autotuner to replicate in the prosthesis. The interface-based approach is validated by observing the ability of the tuning algorithm to successfully converge to various alternate knee profiles by testing on two able-bodied subjects walking with a robotic knee prosthesis. The algorithm was found to converge successfully in an average duration of 1.15 min for the first subject and 2.31 min for the second subject. Further, the subjects displayed different preferences for optimal profiles reinforcing the need to tune alternate profiles. The implications of the results in the tuning of robotic prosthetic devices are discussed.}, journal={2021 IEEE/RSJ INTERNATIONAL CONFERENCE ON INTELLIGENT ROBOTS AND SYSTEMS (IROS)}, publisher={IEEE}, author={Alili, Abbas and Nalam, Varun and Li, Minhan and Liu, Ming and Si, Jennie and Huang, He}, year={2021}, pages={6190–6195} }
 @article{tu_li_liu_si_huang_2020, title={A Data-Driven Reinforcement Learning Solution Framework for Optimal and Adaptive Personalization of a Hip Exoskeleton}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85098609759&partnerID=MN8TOARS}, journal={arXiv}, author={Tu, X. and Li, M. and Liu, M. and Si, J. and Huang, H.H.}, year={2020} }
 @article{liu_kamper_huang_2020, title={An Easy-to-Use Socket-Suspension System Monitor for Lower Limb Amputees}, volume={69}, url={http://dx.doi.org/10.1109/tim.2020.2999738}, DOI={10.1109/tim.2020.2999738}, abstractNote={Prosthetic sockets and the related suspension systems play key roles in functional restoration for individuals with limb loss; a good fit of the socket-suspension systems (SSSs) is critical to avoid unexpected skin breakdown. However, maintaining a good fit between the limb and SSS is challenging due to daily volume changes in the residual limb. In clinical practice, the fit of the SSS is monitored only through haptic feedback from the user. While important, this subjective feedback is often unreliable and inaccurate. Several automatic SSS monitoring approaches have been proposed, but widespread adoption has been limited by related high costs and difficulty of use. To address these issues, we designed a wearable SSS monitor that can be adopted for any SSS using liners without requiring permanent modifications of the SSS. Using a tiny magnetic disk as a marker, we were able to monitor the relative displacement between the residual limb (represented by the liner) and the prosthetic socket along the proximal-distal direction using field orientation measurement sensors. We conducted bench tests to validate the high accuracy of the wearable SSS monitor and demonstrated the capability of tracking SSS fit changes during locomotion without permanent modification of the SSS.}, number={11}, journal={IEEE Transactions on Instrumentation and Measurement}, author={Liu, M. and Kamper, D.G. and Huang, H.}, year={2020}, pages={8973–8982} }
 @article{wu_yao_liu_hu_huang_zhu_2020, title={Buckle-Delamination-Enabled Stretchable Silver Nanowire Conductors}, volume={12}, ISSN={["1944-8252"]}, url={http://dx.doi.org/10.1021/acsami.0c09775}, DOI={10.1021/acsami.0c09775}, abstractNote={Controlled buckling and delamination of thin films on a compliant substrate has attracted much attention for applications ranging from micro/nanofabrication to flexible and stretchable electronics to bioengineering. Here a highly conductive and stretchable conductor is fabricated by attaching a polymer composite film (with a thin layer of silver nanowires embedded below the surface of the polymer matrix) on top of a pre-stretched elastomer substrate followed with releasing the prestrain. A partially delaminated wavy geometry of the polymer film is created. During the evolution of the buckle delamination, the blisters pop up randomly but self-adjust into a uniform distribution, which effectively reduces the local strain in the silver nanowires. The resistance change of the conductor is less than 3% with the applied strain up to 100%. A theoretical model on the buckle-delamination structure is developed to predict the geometrical evolution, which agrees well with experimental observation. Finally, an integrated silver nanowire/elastomer sensing module and a stretchable thermochromic device are developed to demonstrate the utility of the stretchable conductor. This work highlights the important relevance of mechanics-based design in nanomaterial-enabled stretchable devices.}, number={37}, journal={ACS APPLIED MATERIALS & INTERFACES}, publisher={American Chemical Society (ACS)}, author={Wu, Shuang and Yao, Shanshan and Liu, Yuxuan and Hu, Xiaogang and Huang, He Helen and Zhu, Yong}, year={2020}, month={Sep}, pages={41696–41703} }
 @article{fleming_huang_buxton_hodges_huang_2020, title={Direct continuous EMG control of a powered prosthetic ankle for improved postural control after guided physical training: A case study}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85102724227&partnerID=MN8TOARS}, DOI={10.1101/2020.09.11.293373}, abstractNote={AbstractBackgroundDespite the promise of powered lower limb prostheses, the existing control of these modern devices is insufficient to assist many daily activities, such as postural control while lifting weight, that require continuous control of prosthetic joints according to human states and environments. The objective of this case study was to investigate the feasibility and potential of direct, continuous electromyographic (dEMG) control of a powered ankle prosthesis, combined with physical therapist (PT)-guided training, for improved standing postural control in an individual with transtibial amputation.MethodsA powered prosthetic ankle was directly controlled by EMG signals of the residual lateral gastrocnemius and tibialis anterior muscles. The participant with transtibial amputation received 4-week PT-guided training on posture while using the dEMG control of powered ankle. A subset of activities in the mini-BESTest (a clinical balance assessment tool) were used in the training and evaluation protocol. We quantified EMG signals in the bilateral shank muscles, biomechanics that captures postural control and stability, and score for the clinical balance evaluation.ResultsCompared to the participant’s daily passive prosthesis, the dEMG-controlled ankle, combined with the training, yielded improved clinical balance score and reduced compensation from the intact joints. In addition, cross correlation coefficient of bilateral CoP excursions, a metric for quantifying standing postural control, increased to 0.83(±0.07) when using dEMG ankle control, compared with 0.39(±0.29) when using the passive device. Between-limb coordination was also observed as synchronized activation of homologous muscles in the shank. We witnessed rapid improvement in performance on the first day of the training for load transfer tasks, where bilateral CoP synchronization improvement was significantly related to repetition order (R=0.459, p = 0.045). Finally, the participant further improved this performance significantly across training days.ConclusionThis case study showed the feasibility of dEMG control of powered prosthetic ankle by a transtibial amputee after a PT-guided training to assist postural control. This study’s training protocol and dEMG control method that lays the foundation for future study to extend these results through the inclusion of more participants and activities.}, journal={bioRxiv}, author={Fleming, A. and Huang, S. and Buxton, E. and Hodges, F. and Huang, H.}, year={2020} }
 @inproceedings{park_zahabi_kaber_ruiz_huang_2020, title={Evaluation of Activities of Daily Living Tesbeds for Assessing Prosthetic Device Usability}, url={http://dx.doi.org/10.1109/ichms49158.2020.9209553}, DOI={10.1109/ichms49158.2020.9209553}, abstractNote={Individuals with upper limb amputations rely on prosthetic devices to perform activities of daily living. However, these technologies are often reported as being difficult to use. Prior studies have used a variety of testbeds to assess the usability of prosthetic devices. However, there is no defined strategy for selecting the most effective batteries. This study developed a task selection strategy and a test battery to assess usability of upper-limb prosthetic devices. A combination of methods was applied, including a constrained literature review, sensitivity analysis, and review of fundamental upper-limb movements. Findings suggest that the clothespin relocation task (CRT) and Southampton hand assessment protocol (SHAP) are the most sensitive testbeds for usability assessment of upper-limb prosthetic devices and these tasks require similar limb movements to high frequency ADLs.}, booktitle={2020 IEEE International Conference on Human-Machine Systems (ICHMS)}, publisher={IEEE}, author={Park, Junho and Zahabi, Maryam and Kaber, David and Ruiz, Jaime and Huang, He Helen}, year={2020}, month={Sep} }
 @article{silva_starliper_zhong_huang_lobaton_2020, title={Evaluation of embedded platforms for lower limb prosthesis with visual sensing capabilities}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85095221966&partnerID=MN8TOARS}, journal={arXiv}, author={Silva, R.L. and Starliper, N. and Zhong, B. and Huang, H.H. and Lobaton, E.}, year={2020} }
 @article{pan_vargas_vargas_fleming_fleming_hu_hu_zhu_huang_2020, title={Evoking haptic sensations in the foot through high-density transcutaneous electrical nerve stimulations}, volume={4}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85086524144&partnerID=MN8TOARS}, DOI={10.1088/1741-2552/ab8e8d}, abstractNote={Objective. Evoking haptic sensation on upper limb amputees via peripheral nerve stimulation has been investigated intensively in the past decade, but related studies involving lower limb amputees are limited. This study aimed to evaluate the feasibility of using non-invasive transcutaneous electrical nerve stimulation to evoke haptic sensation along the phantom limb of the amputated foot of transtibial amputees. Approach. A high-density electrode grid (4 × 4) was placed over the skin surface above the distal branching of the sciatic, tibial, and common peroneal nerves. We hypothesized that electrical stimulation delivered to distinct electrode pairs created unique electric fields, which can activate selective sets of sensory axons innervating different skin regions of the foot. Five transtibial amputee subjects (three unilateral and two bilateral) and one able-bodied subject were tested by scanning all possible electrode pair combinations. Main results. All subjects reported various haptic percepts at distinct regions along the foot with each corresponding to specific electrode pairs. These results demonstrated the capability of our non-invasive nerve stimulation method to evoke haptic sensations in the foot of transtibial amputees and the able-bodied subject. Significance. The outcomes contribute important knowledge and evidence regarding missing tactile sensation in the foot of lower limb amputees and might also facilitate future development of strategies to manage phantom pain and enhance embodiment of prosthetic legs in the future.}, number={3}, journal={Journal of Neural Engineering}, publisher={IOP Publishing}, author={Pan, L. and Vargas, L. and Vargas, L. and Fleming, A. and Fleming, A. and Hu, X. and Hu, X. and Zhu, Y. and Huang, H.}, year={2020}, month={Apr} }
 @article{mcknight_tabor_agcayazi_fleming_ghosh_huang_bozkurt_2020, title={Fully-textile Insole Seam-line for Multi-modal Sensor Mapping}, volume={20}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85089489688&partnerID=MN8TOARS}, DOI={10.1109/jsen.2020.2990627}, abstractNote={Here we demonstrate a fully-textile insole sensor capable of providing simultaneous pressure and wetness sensing. By utilizing a three-layer textile structure consisting of knit/non-woven materials and conductive yarns, we produced a pressure sensitive sensing array. An additional layer of embedded conductive yarns enabled mapping of wetness in the insole environment. The pressure sensing array was capable of detecting pressures up to 600 kPa, with enhanced sensitivity in the lower pressure regime (0-150 kPa) and was found to be stable to static/dynamic loading conditions following calibration. The system was demonstrated as an insole sensor for mapping plantar pressures during normal gait walking. As a wetness sensor, the system was capable of characterizing location and extent of wetness as well as distinguishing between ionic/non-ionic wetting fluids. We demonstrated detection of as little as $50~\mu \text{L}$ of saline solution at a single wetness sensing point. This fully-textile design provides conformal, breathable sensing for mapping pressures and wetness across curvilinear surfaces. This design enables spatial/temporal resolution of these parameters and could provide a method for long-term monitoring of dynamic skin interface environments to provide improved customized orthotic/orthopedic solutions.}, number={17}, journal={IEEE Sensors Journal}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Mcknight, Michael and Tabor, Jordan and Agcayazi, Talha and Fleming, Aaron and Ghosh, Tushar and Huang, He and Bozkurt, Alper}, year={2020}, pages={1–1} }
 @article{fylstra_lee_huang_brandt_lewek_huang_2020, title={Human-prosthesis coordination: A preliminary study exploring coordination with a powered ankle-foot prosthesis}, volume={80}, url={http://dx.doi.org/10.1016/j.clinbiomech.2020.105171}, DOI={10.1016/j.clinbiomech.2020.105171}, abstractNote={Powered ankle-foot prostheses were developed to replicate the mechanics of the biological ankle by providing positive work during the push-off phase of gait. However, the benefits of powered prostheses on improving overall human gait efficiency (usually quantified by metabolic cost) have not been consistently shown. Here, we have focused on the mechanical work produced at the prosthetic ankle and its interaction with the amputee's movement.Five unilateral transtibial amputees walked on a treadmill using 1) a powered ankle-foot prosthesis and 2) their daily passive device. We determined the net ankle work and ankle work loops on the prosthesis-side to quantify the efficiency of the human-prosthesis physical interaction. We further studied peak propulsion timing and the posture of the amputee's lower limb and prosthesis as indicators of the human-prosthesis coordination. Comparisons were made between the passive and powered prosthesis conditions for each participant.The powered prosthesis did not consistently increase net ankle work compared to each participant's passive device. For participants that lacked efficiency in interacting with the powered prosthesis, we observed 1) early prosthesis-side peak propulsion timing (≥ 4% earlier) and 2) a more vertical residual shank at the time of peak propulsion (> 2° more vertical) indicating that the human's limb movement and the prosthesis control during push-off were not well coordinated.Results from this preliminary study highlight the need for future work to systematically quantify the coordination between the human and powered prosthesis and understand how such coordination at the joint level influences overall gait efficiency.}, journal={Clinical Biomechanics}, publisher={Elsevier BV}, author={Fylstra, B.L. and Lee, I.-C. and Huang, S. and Brandt, A. and Lewek, M.D. and Huang, H.H.}, year={2020}, month={Dec}, pages={105171} }
 @inproceedings{gao_si_wen_li_huang_2020, title={Knowledge-Guided Reinforcement Learning Control for Robotic Lower Limb Prosthesis}, url={http://dx.doi.org/10.1109/icra40945.2020.9196749}, DOI={10.1109/icra40945.2020.9196749}, abstractNote={Robotic prostheses provide new opportunities to better restore lost functions than passive prostheses for trans-femoral amputees. But controlling a prosthesis device automatically for individual users in different task environments is an unsolved problem. Reinforcement learning (RL) is a naturally promising tool. For prosthesis control with a user in the loop, it is desirable that the controlled prosthesis can adapt to different task environments as quickly and smoothly as possible. However, most RL agents learn or relearn from scratch when the environment changes. To address this issue, we propose the knowledge-guided Q-learning (KG-QL) control method as a principled way for the problem. In this report, we collected and used data from two able-bodied (AB) subjects wearing a RL controlled robotic prosthetic limb walking on level ground. Our ultimate goal is to build an efficient RL controller with reduced time and data requirements and transfer knowledge from AB subjects to amputee subjects. Toward this goal, we demonstrate its feasibility by employing OpenSim, a well-established human locomotion simulator. Our results show the OpenSim simulated amputee subject improved control tuning performance over learning from scratch by utilizing knowledge transfer from AB subjects. Also in this paper, we will explore the possibility of information transfer from AB subjects to help tuning for the amputee subjects.}, booktitle={2020 IEEE International Conference on Robotics and Automation (ICRA)}, publisher={IEEE}, author={Gao, Xiang and Si, Jennie and Wen, Yue and Li, Minhan and Huang, He Helen}, year={2020}, month={May}, pages={754–760} }
 @article{stallrich_islam_staicu_crouch_pan_huang_2020, title={OPTIMAL EMG PLACEMENT FOR A ROBOTIC PROSTHESIS CONTROLLER WITH SEQUENTIAL, ADAPTIVE FUNCTIONAL ESTIMATION (SAFE)}, volume={14}, ISSN={["1932-6157"]}, url={http://dx.doi.org/10.1214/20-aoas1324}, DOI={10.1214/20-AOAS1324}, abstractNote={Robotic hand prostheses require a controller to decode muscle contraction information, such as electromyogram (EMG) signals, into the user’s desired hand movement. State-of-the-art decoders demand extensive training, require data from a large number of EMG sensors, and are prone to poor predictions. Biomechanical models of a single movement degree-of-freedom tell us that relatively few muscles, and hence fewer EMG sensors, are needed to predict movement. We propose a novel decoder based on a dynamic, functional linear model with velocity or acceleration as its response and the recent past EMG signals as functional covariates. The effect of each EMG signal varies with the recent position to account for biomechanical features of hand movement, increasing the predictive capability of a single EMG signal compared to existing decoders. The effects are estimated with a multi-stage, adaptive estimation procedure we call Sequential Adaptive Functional Estimation (SAFE). Starting with 16 potential EMG sensors, our method correctly identifies the few EMG signals that are known to be important for an able-bodied subject. Furthermore, the estimated effects are interpretable and can significantly improve understanding and development of robotic hand prostheses.}, number={3}, journal={ANNALS OF APPLIED STATISTICS}, publisher={Institute of Mathematical Statistics}, author={Stallrich, Jonathan and Islam, Md Nazmul and Staicu, Ana-Maria and Crouch, Dustin and Pan, Lizhi and Huang, He}, year={2020}, month={Sep}, pages={1164–1181} }
 @article{vargas_huang_zhu_hu_2020, title={Object Shape and Surface Topology Recognition Using Tactile Feedback Evoked through Transcutaneous Nerve Stimulation}, volume={13}, ISSN={["2329-4051"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85078214377&partnerID=MN8TOARS}, DOI={10.1109/TOH.2020.2967366}, abstractNote={Tactile feedback is critical for distinguishing different object properties. In this article, we determined if tactile feedback evoked by transcutaneous nerve stimulation can be used to detect objects of different shape and surface topology. To evoke tactile sensation at different fingers, a 2x8 electrode grid was placed along the subject's upper arm, and two concurrent electrical stimulation trains targeted the median and ulnar nerve bundles, which evoked individually modulated sensations at different fingers. Fingertip forces of the prosthetic hand were transformed to stimulation current amplitude. Object shape was encoded based on finger-object contact timing. Surface topology represented by ridge height and spacing was encoded through current amplitude and stimulation time interval, respectively. The elicited sensation allowed subjects to determine object shape with success rates >84%. Surface topology recognition resulted in success rates >81%. Our findings suggest that tactile feedback evoked from transcutaneous nerve stimulation allows the recognition of object shape and surface topology. The ability to recognize these properties may help improve object manipulation and promote fine control of a prosthetic hand.}, number={1}, journal={IEEE TRANSACTIONS ON HAPTICS}, author={Vargas, Luis and Huang, He and Zhu, Yong and Hu, Xiaogang}, year={2020}, pages={152–158} }
 @article{lee_pacheco_lewek_huang_2020, title={Perceiving amputee gait from biological motion: kinematics cues and effect of experience level}, volume={10}, url={http://dx.doi.org/10.1038/s41598-020-73838-y}, DOI={10.1038/s41598-020-73838-y}, abstractNote={AbstractPhysical therapists (PT) and clinicians must be skilled in identifying gait features through observation to assess motor deficits in patients and intervene appropriately. Inconsistent results in the literature have led researchers to question how clinical experience influences PT’s gait perception and to seek the key kinematic features that should be trained to enhance PT’s skill. Thus, this study investigated (1) what are the informative kinematic features that allow gait-deviation perception in amputee gait and (2) whether there are differences in observational gait skills between PT and individuals with less clinical experience (PT students [PTS] and Novices). We introduced a new method that combines biological motion and principal component analysis to gradually mesh amputee and typical walking patterns. Our analysis showed that on average the accuracy rate in identifying gait deviations between PT and PTS was similar and better than Novices. Also, we found that PT’s experience was demonstrated by their better perception of gait asymmetry. The extracted principal components demonstrated that the major gait deviation of amputees was the medial–lateral body sway and spatial gait asymmetry.}, number={1}, journal={Scientific Reports}, publisher={Springer Science and Business Media LLC}, author={Lee, I.-Chieh and Pacheco, Matheus M. and Lewek, Michael D. and Huang, He}, year={2020}, month={Dec} }
 @article{gao_si_wen_li_huang_2020, title={Reinforcement learning control of robotic knee with human in the loop by flexible policy iteration}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85094999958&partnerID=MN8TOARS}, journal={arXiv}, author={Gao, X. and Si, J. and Wen, Y. and Li, M. and Huang, H.}, year={2020} }
 @inproceedings{vargas_huang_zhu_hu_2020, title={Stiffness Perception using Transcutaneous Electrical Stimulation during Active and Passive Prosthetic Control}, volume={2020-July}, url={http://dx.doi.org/10.1109/embc44109.2020.9176078}, DOI={10.1109/embc44109.2020.9176078}, abstractNote={Haptic feedback allows an individual to identify various object properties. In this preliminary study, we determined the performance of stiffness recognition using transcutaneous nerve stimulation when a prosthetic hand was moved passively or was controlled actively by the subjects. Using a 2x8 electrode grid placed along the subject’s upper arm, electrical stimulation was delivered to evoke somatotopic sensation along their index finger. Stimulation intensity, i.e. sensation strength, was modulated using the fingertip forces from a sensorized prosthetic hand. Object stiffness was encoded based on the rate of change of the evoked sensation as the prosthesis grasped one of three objects of different stiffness levels. During active control, sensation was modulated in real time as recorded forces were converted to stimulation amplitudes. During passive control, prerecorded force traces were randomly selected from a pool. Our results showed that the accuracy of object stiffness recognition was similar in both active and passive conditions. A slightly lower accuracy was observed during active control in one subject, which indicated that the sensorimotor integration processes could affect haptic perception for some users.}, booktitle={2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)}, publisher={IEEE}, author={Vargas, Luis and Huang, Helen and Zhu, Yong and Hu, Xiaogang}, year={2020}, month={Jul}, pages={3909–3912} }
 @article{li_wen_gao_si_huang_2020, title={Towards expedited impedance tuning of a robotic prosthesis for personalized gait assistance by reinforcement learning control}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85110372419&partnerID=MN8TOARS}, journal={arXiv}, author={Li, M. and Wen, Y. and Gao, X. and Si, J. and Huang, H.}, year={2020} }
 @article{wu_saul_huang_2021, title={Using Reinforcement Learning to Estimate Human Joint Moments From Electromyography or Joint Kinematics: An Alternative Solution to Musculoskeletal-Based Biomechanics}, volume={143}, ISBN={1528-8951}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85107163923&partnerID=MN8TOARS}, DOI={10.1115/1.4049333}, abstractNote={AbstractReinforcement learning (RL) has potential to provide innovative solutions to existing challenges in estimating joint moments in motion analysis, such as kinematic or electromyography (EMG) noise and unknown model parameters. Here, we explore feasibility of RL to assist joint moment estimation for biomechanical applications. Forearm and hand kinematics and forearm EMGs from four muscles during free finger and wrist movement were collected from six healthy subjects. Using the proximal policy optimization approach, we trained two types of RL agents that estimated joint moment based on measured kinematics or measured EMGs, respectively. To quantify the performance of trained RL agents, the estimated joint moment was used to drive a forward dynamic model for estimating kinematics, which was then compared with measured kinematics using Pearson correlation coefficient. The results demonstrated that both trained RL agents are feasible to estimate joint moment for wrist and metacarpophalangeal (MCP) joint motion prediction. The correlation coefficients between predicted and measured kinematics, derived from the kinematics-driven agent and subject-specific EMG-driven agents, were 98% ± 1% and 94% ± 3% for the wrist, respectively, and were 95% ± 2% and 84% ± 6% for the metacarpophalangeal joint, respectively. In addition, a biomechanically reasonable joint moment-angle-EMG relationship (i.e., dependence of joint moment on joint angle and EMG) was predicted using only 15 s of collected data. In conclusion, this study illustrates that an RL approach can be an alternative technique to conventional inverse dynamic analysis in human biomechanics study and EMG-driven human-machine interfacing applications.}, number={4}, journal={JOURNAL OF BIOMECHANICAL ENGINEERING-TRANSACTIONS OF THE ASME}, author={Wu, Wen and Saul, Katherine R. and Huang, He}, year={2021} }
 @article{wen_li_si_huang_2020, title={Wearer-Prosthesis Interaction for Symmetrical Gait: A Study Enabled by Reinforcement Learning Prosthesis Control}, volume={28}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85083163810&partnerID=MN8TOARS}, DOI={10.1109/TNSRE.2020.2979033}, abstractNote={With advances in robotic prostheses, rese-archers attempt to improve amputee’s gait performance (e.g., gait symmetry) beyond restoring normative knee kinematics/kinetics. Yet, little is known about how the prosthesis mechanics/control influence wearer-prosthesis’ gait performance, such as gait symmetry, stability, etc. This study aimed to investigate the influence of robotic transfemoral prosthesis mechanics on human wearers’ gait symmetry. The investigation was enabled by our previously designed reinforcement learning (RL) supplementary control, which simultaneously tuned 12 control parameters that determined the prosthesis mechanics throughout a gait cycle. The RL control design facilitated safe explorations of prosthesis mechanics with the human in the loop. Subjects were recruited and walked with a robotic transfemoral prosthesis on a treadmill while the RL controller tuned the control parameters. Stance time symmetry, step length symmetry, and bilateral anteroposterior (AP) impulses were measured. The data analysis showed that changes in robotic knee mechanics led to movement variations in both lower limbs and therefore gait temporal-spatial symmetry measures. Consistent across all the subjects, inter-limb AP impulse measurements explained gait symmetry: the stance time symmetry was significantly correlated with the net inter-limb AP impulse, and the step length symmetry was significantly correlated with braking and propulsive impulse symmetry. The results suggest that it is possible to personalize transfemoral prosthesis control for improved temporal-spatial gait symmetry. However, adjusting prosthesis mechanics alone was insufficient to maximize the gait symmetry. Rather, achieving gait symmetry may require coordination between the wearer’s motor control of the intact limb and adaptive control of the prosthetic joints. The results also indicated that the RL-based prosthesis tuning system was a potential tool for studying wearer-prosthesis interactions.}, number={4}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Wen, Yue and Li, Minhan and Si, Jennie and Huang, He}, year={2020}, month={Apr}, pages={904–913} }
 @article{zhang_tran_huang_2019, title={Admittance Shaping-Based Assistive Control of SEA-Driven Robotic Hip Exoskeleton}, volume={24}, ISSN={["1941-014X"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85085177698&partnerID=MN8TOARS}, DOI={10.1109/TMECH.2019.2916546}, abstractNote={This paper presents an admittance shaping-based assistive control for a series elastic actuator (SEA) driven robotic hip exoskeleton that can assist individuals with hip muscle weakness to restore normative mobility. The motivation for this paper is to develop a unified controller framework for designing an SEA-driven hip exoskeleton to assist walking and enhance gait stability. The controller design aims to modify the dynamic response of a coupled human-exoskeleton system, i.e., the relationship between the net muscle torque exerted by the human and the resulting angular motion, to ensure strong human-exoskeleton synergy to provide the effective assistance. This controller was preliminarily evaluated on a healthy subject walking on a treadmill at a speed of 1.0 m/s. Results showed that the exoskeleton can effectively provide walking assistance to the human by reducing electromyography (EMG) activation and increasing agility during locomotion. Specifically, EMG was reduced 3.3%–38% when walking with the hip exoskeleton when compared to walking without wearing the hip exoskeleton. In addition, timing of the maximum hip flexion angle increased by 10% (moved from 42% to 32% of gait cycle) when the controller had an inertia compensation of 60%. The faster onset of the maximum flexion angle will allow the wearer to more quickly generate reactive steps when trying to avoid a fall. Future work will aim to apply the hip exoskeleton to persons having hip muscle weakness or other musculoskeletal impairment, to restore hip movement and enough hip force to walk normally.}, number={4}, journal={IEEE-ASME TRANSACTIONS ON MECHATRONICS}, author={Zhang, Ting and Tran, Minh and Huang, He}, year={2019}, month={Aug}, pages={1508–1519} }
 @article{zahabi_white_zhang_winslow_zhang_huang_kaber_2019, title={Application of Cognitive Task Performance Modeling for Assessing Usability of Transradial Prostheses}, volume={49}, ISSN={["2168-2305"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85063397939&partnerID=MN8TOARS}, DOI={10.1109/THMS.2019.2903188}, abstractNote={The goal of this study was to investigate the use of cognitive modeling to assess the usability of an upper-limb prosthesis with a focus on mental workload responses. Prior studies have investigated usability of upper-limb prostheses with subjective surveys and physiological measures. However, these approaches have limitations, including subject recall of conditions and physiological response contamination by head and body movements and user speech during task performance as well as sensitivity to physical fatigue and room lighting conditions. Cognitive modeling was used to assess mental workload in use of transradial upper-limb prosthesis. A case study was conducted with a participant with upper-limb amputation using two different types of electromyography-based control schemes, including conventional direct control (DC) and pattern recognition (PR) control in order to compare cognitive model outcomes with mental workload assessment using eye-tracking measures. Cognitive models time estimates were also compared with actual task completion time results from the case study to further assess the validity of cognitive modeling as an analytical tool for evaluating upper limb prosthesis usability. Findings of both the cognitive models and case study revealed the PR mode to be more intuitive, reduce cognitive load, and increase efficiency in prosthetic control as compared to the DC mode. Results of the present study revealed that cognitive modeling can be used as an analytical approach for assessing upper-limb prosthetic device usability in terms of workload outcomes. Future studies should validate the present findings with more precise time estimations and a larger user sample size.}, number={4}, journal={IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS}, author={Zahabi, Maryam and White, Melissa Mae and Zhang, Wenjuan and Winslow, Anna T. and Zhang, Fan and Huang, He and Kaber, David B.}, year={2019}, month={Aug}, pages={381–387} }
 @article{pan_crouch_huang_2019, title={Comparing EMG-Based Human-Machine Interfaces for Estimating Continuous, Coordinated Movements}, volume={27}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85073667144&partnerID=MN8TOARS}, DOI={10.1109/TNSRE.2019.2937929}, abstractNote={Electromyography (EMG)-based interfaces are trending toward continuous, simultaneous control with multiple degrees of freedom. Emerging methods range from data-driven approaches to biomechanical model-based methods. However, there has been no direct comparison between these two types of continuous EMG-based interfaces. The aim of this study was to compare a musculoskeletal model (MM) with two data-driven approaches, linear regression (LR) and artificial neural network (ANN), for predicting continuous wrist and hand motions for EMG-based interfaces. Six able-bodied subjects and one transradial amputee subject performed (missing) metacarpophalangeal (MCP) and wrist flexion/extension, simultaneously or independently, while four EMG signals were recorded from forearm muscles. To add variation to the EMG signals, the subjects repeated the MCP and wrist motions at various upper extremity postures. For each subject, the EMG signals collected from the neutral posture were used to build the EMG interfaces; the EMG signals collected from all postures were used to evaluate the interfaces. The performance of the interface was quantified by Pearson’s correlation coefficient (r) and the normalized root mean square error (NRMSE) between measured and estimated joint angles. The results demonstrated that the MM predicted movements more accurately, with higher r values and lower NRMSE, than either LR or ANN. Similar results were observed in the transradial amputee. Additionally, the variation in r across postures, an indicator of reliability against posture changes, was significantly lower (better) for the MM than for either LR or ANN. Our findings suggest that incorporating musculoskeletal knowledge into EMG-based human-machine interfaces could improve the estimation of continuous, coordinated motion.}, number={10}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Pan, Lizhi and Crouch, Dustin L. and Huang, He}, year={2019}, month={Oct}, pages={2145–2154} }
 @article{zhang_huang_2019, title={Design and Control of a Series Elastic Actuator With Clutch for Hip Exoskeleton for Precise Assistive Magnitude and Timing Control and Improved Mechanical Safety}, volume={24}, ISSN={1083-4435 1941-014X}, url={http://dx.doi.org/10.1109/tmech.2019.2932312}, DOI={10.1109/TMECH.2019.2932312}, abstractNote={Transparency and guaranteed safety are important requirements in the design of wearable exoskeleton actuators for individuals who have lower limb deficits but still maintain a certain level of voluntary motor control. Specifically, precision in torque delivery timing and magnitude, robustness, disturbance rejection, and repeatability are desired in the actuator design and control. Motivated by these needs, this study aims to develop a series of elastic actuators with clutch (SEAC) that can precisely generate the desired assistance in terms of both timing and torque magnitude for a wearable hip exoskeleton and guarantee the wearer's safety at the same time. The proposed mechanical design improves actuator transparency and safety by a mechanical clutch that automatically disengages the transmission when needed. A new torque control for the SEAC, based on singular perturbation theory with flexible compensation techniques, is proposed to precisely control the assistive torque by rejecting the undesired human motion disturbance. The mechanical design of the proposed device and the design of a singular perturbation control algorithm are discussed, and the SEAC performance is verified by experiments. Experimental results, derived from a test with a human subject, are presented to demonstrate the precision of the assistive torque and timing control of the SEAC while interacting with a human wearer.}, number={5}, journal={IEEE/ASME Transactions on Mechatronics}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Zhang, Ting and Huang, He}, year={2019}, month={Oct}, pages={2215–2226} }
 @article{brandt_riddick_stallrich_lewek_huang_2019, title={Effects of extended powered knee prosthesis stance time via visual feedback on gait symmetry of individuals with unilateral amputation: a preliminary study}, volume={16}, ISSN={["1743-0003"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85072172371&partnerID=MN8TOARS}, DOI={10.1186/s12984-019-0583-z}, abstractNote={AbstractBackgroundEstablishing gait symmetry is a major aim of amputee rehabilitation and may be more attainable with powered prostheses. Though, based on previous work, we postulate that users transfer a previously-learned motor pattern across devices, limiting the functionality of more advanced prostheses. The objective of this study was to preliminarily investigate the effect of increased stance time via visual feedback on amputees’ gait symmetry using powered and passive knee prostheses.MethodsFive individuals with transfemoral amputation or knee disarticulation walked at their self-selected speed on a treadmill. Visual feedback was used to promote an increase in the amputated-limb stance time. Individuals were fit with a commercially-available powered prosthesis by a certified prosthetist and practiced walking during a prior visit. The same protocol was completed with a passive knee and powered knee prosthesis on separate days. We used repeated-measures, two-way ANOVA (alpha = 0.05) to test for significant effects of the feedback and device factors. Our main outcome measures were stance time asymmetry, peak anterior-posterior ground reaction forces, and peak anterior propulsion asymmetry.ResultsIncreasing the amputated-limb stance time via visual feedback significantly improved the stance time symmetry (p = 0.012) and peak propulsion symmetry (p = 0.036) of individuals walking with both prostheses. With the powered knee prosthesis, the highest feedback target elicited 36% improvement in stance time symmetry, 22% increase in prosthesis-side peak propulsion, and 47% improvement in peak propulsion symmetry compared to a no feedback condition. The changes with feedback were not different with the passive prosthesis, and the main effects of device/ prosthesis type were not statistically different. However, subject by device interactions were significant, indicating individuals did not respond consistently with each device (e.g. prosthesis-side propulsion remained comparable to or was greater with the powered versus passive prosthesis for different subjects). Overall, prosthesis-side peak propulsion averaged across conditions was 31% greater with the powered prosthesis and peak propulsion asymmetry improved by 48% with the powered prosthesis.ConclusionsIncreasing prosthesis-side stance time via visual feedback favorably improved individuals’ temporal and propulsive symmetry. The powered prosthesis commonly enabled greater propulsion, but individuals adapted to each device with varying behavior, requiring further investigation.}, number={1}, journal={JOURNAL OF NEUROENGINEERING AND REHABILITATION}, author={Brandt, Andrea and Riddick, William and Stallrich, Jonathan and Lewek, Michael and Huang, He Helen}, year={2019}, month={Sep} }
 @article{brandt_huang_2019, title={Effects of extended stance time on a powered knee prosthesis and gait symmetry on the lateral control of balance during walking in individuals with unilateral amputation}, volume={16}, ISSN={["1743-0003"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85075794161&partnerID=MN8TOARS}, DOI={10.1186/s12984-019-0625-6}, abstractNote={Abstract
                Background
                Individuals with lower limb amputation commonly exhibit large gait asymmetries that are associated with secondary health issues. It has been shown that they are capable of attaining improved temporal and propulsive symmetry when walking with a powered knee prosthesis and visual feedback, but they perceive this pattern of gait to be more difficult. Rather than improving the efficiency of gait, improved gait symmetry may be increasing individuals’ effort associated with maintaining lateral balance.
              
                Methods
                In this study, we used a simple visual feedback paradigm to increase the prosthesis-side stance time of six individuals with unilateral TFA or KD as they walked on a powered knee prosthesis at their self-selected speed. As they walked more symmetrically, we evaluated changes in medial-lateral center-of-mass excursion, lateral margin of stability, stride width, and hip abductor activity.
              
                Results
                As the subjects increased their prosthesis-side stance time, their center-of-mass excursion and hip abductor activity significantly increased, while their lateral margin of stability significantly decreased on the prosthesis-side only. Stride width remained relatively unchanged with testing condition.
              
                Conclusions
                Extended stance time on a powered knee prosthesis (yielding more symmetric gait) challenged the lateral balance of individuals with lower limb amputation. Lateral stability may be a reason they prefer an asymmetric gait, even with more advanced technology. Hip muscular changes post-amputation may contribute to the decline in stability on the prosthesis side. Interventions and advancements in prosthesis control aimed at improving their control of lateral balance may ameliorate the difficulty in walking with improved gait symmetry.
              }, number={1}, journal={JOURNAL OF NEUROENGINEERING AND REHABILITATION}, author={Brandt, Andrea and Huang, He}, year={2019}, month={Nov} }
 @article{qin_li_yao_liu_huang_zhu_2019, title={Electrocardiogram of a Silver Nanowire Based Dry Electrode: Quantitative Comparison With the Standard Ag/AgCl Gel Electrode}, volume={7}, ISSN={["2169-3536"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85062896682&partnerID=MN8TOARS}, DOI={10.1109/ACCESS.2019.2897590}, abstractNote={Novel dry electrodes have promoted the development of wearable electrocardiogram (ECG) that is collected in daily life to monitor the ambulatory activity of heart status. To evaluate the performance of a dry electrode, it is necessary to compare it with the commercial disposable silver/silver chloride (Ag/AgCl) gel electrode. In this paper, a silver nanowire (AgNW)-based dry electrode was fabricated for noninvasive and wearable ECG sensing. Signals from the AgNW electrode and the Ag/AgCl electrode were simultaneously collected in two conditions: sitting and walking. Signal quality was evaluated in terms of ECG morphology, R-peak to R-peak interval, and heart rate variability analysis. Quantitative comparisons showed that the AgNW electrode could collect acceptable ECG waveforms as the Ag/AgCl electrode in both the sitting and walking conditions. However, the baseline drift and waveform distortions existed in the AgNW electrode, likely due to electrode motion. If the skin-electrode contact is improved, the dry electrode can be a promising substitute for the Ag/AgCl electrode.}, journal={IEEE ACCESS}, author={Qin, Qin and Li, Jianqing and Yao, Shanshan and Liu, Chengyu and Huang, He and Zhu, Yong}, year={2019}, pages={20789–20800} }
 @article{vargas_whitehouse_huang_zhu_hu_2019, title={Evoked Haptic Sensation in the Hand With Concurrent Non-Invasive Nerve Stimulation}, volume={66}, ISSN={["1558-2531"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85077396250&partnerID=MN8TOARS}, DOI={10.1109/TBME.2019.2895575}, abstractNote={Objective: Haptic perception is critical for prosthetic users to control their prosthetic hand intuitively. In this study, we seek to evaluate the haptic perception evoked from concurrent stimulation trains through multiple channels using transcutaneous nerve stimulation. Methods: A 2 × 8 electrode grid was used to deliver current to the median and ulnar nerves in the upper arm. Different electrodes were first selected to activate the sensory axons, which can elicit sensations at different locations of the hand. Charge-balanced bipolar stimulation was then delivered to two sets of electrodes concurrently with a phase delay (dual stimulation) to determine whether the evoked sensation can be constructed from sensations of single stimulation delivered separately at different locations (single stimulation) along the electrode grid. The temporal delay between the two stimulation trains was altered to evaluate potential interference. The short-term stability of the haptic sensation within a testing session was also evaluated. Results: The evoked sensation during dual stimulation was largely a direct summation of the sensation from single stimulations. The delay between the two stimulation locations had minimal effect on the evoked sensations, which was also stable over repeated testing within a session. Conclusion: Our results indicated that the haptic sensations at different regions of the hand can be constructed by combining the response from multiple stimulation trains directly. The interference between stimulations were minimal. Significance: The outcomes will allow us to construct specific haptic sensation patterns when the prosthesis interacts with different objects, which may help improve user embodiment of the prosthesis.}, number={10}, journal={IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING}, author={Vargas, Luis and Whitehouse, Graham and Huang, He and Zhu, Yong and Hu, Xiaogang}, year={2019}, month={Oct}, pages={2761–2767} }
 @inproceedings{li_zhong_liu_lee_fylstra_lobaton_huang_2019, title={Gaze Fixation Comparisons Between Amputees and Able-bodied Individuals in Approaching Stairs and Level-ground Transitions: A Pilot Study}, ISBN={9781538613115}, url={http://dx.doi.org/10.1109/embc.2019.8857388}, DOI={10.1109/embc.2019.8857388}, abstractNote={This paper aims to investigate the visual strategy of transtibial amputees while they are approaching the transition between level-ground and stairs and compare it with that of able-bodied individuals. To this end, we conducted a pilot study where two transtibial amputee subjects and two able-bodied subjects transitioned from level-ground to stairs and vice versa while wearing eye tracking glasses to record gaze fixations. To investigate how vision functioned to both populations for preparing locomotion on new terrains, gaze fixation behavior before the new terrains were analyzed and compared between two populations across all transition cases in the study. Our results presented that, unlike the able-bodied population, amputees had most of their fixations directed on the transition region prior to new terrains. Furthermore, amputees showed an increased need for visual information during transition regions before navigation on stairs than that before navigation onto level-ground. The insights about amputees’ visual behavior gained by the study may lead the future development of technologies related to the intention prediction and the locomotion recognition for amputees.}, booktitle={2019 41st Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Li, Minhan and Zhong, Boxuan and Liu, Ziwei and Lee, I-Chieh and Fylstra, Bretta L. and Lobaton, Edgar and Huang, He Helen}, year={2019}, month={Jul}, pages={3163–3166} }
 @inproceedings{liu_lupiani_lee_huang_2019, title={Identify Kinematic Features for Powered Prosthesis Tuning}, volume={2019-June}, ISBN={9781728127552}, url={http://dx.doi.org/10.1109/icorr.2019.8779516}, DOI={10.1109/icorr.2019.8779516}, abstractNote={To maximize the benefits of the newly developed powered prosthetic legs, amputees must rely on tuning experts (TE) from manufacturers to tune these devices based on their specific physical conditions. Because TEs are hard to train, it is difficult to access the TEs and the cost of customization is high. If the knowledge used by the TEs could be extracted, it is possible to reduce the tuning cost by automating the tuning procedure or developing efficient TE training programs. In this paper, we preliminarily identified kinematic features that are sensitive to the control parameter change of the powered prosthetic leg. Using data collected from three transtibial amputee subjects with four levels of push-off power, we tested whether a change of push-off power could generate a significant difference on 13 preselected kinematic features during level ground walking at self-selected walking speed. Six features across three joints on the prosthesis side were demonstrated to be sensitive to the change of push-off power.}, booktitle={2019 IEEE 16th International Conference on Rehabilitation Robotics (ICORR)}, publisher={IEEE}, author={Liu, Ming and Lupiani, Ashling and Lee, I-Chieh and Huang, He Helen}, year={2019}, month={Jun}, pages={565–569} }
 @article{vargas_shin_huang_zhu_hu_2019, title={Object stiffness recognition using haptic feedback delivered through transcutaneous proximal nerve stimulation}, volume={17}, ISSN={1741-2552}, url={http://dx.doi.org/10.1088/1741-2552/ab4d99}, DOI={10.1088/1741-2552/ab4d99}, abstractNote={Objective. Haptic feedback is crucial when we manipulate objects. Information pertaining to an object’s stiffness in particular can help facilitate fine motor control. In this study, we seek to determine whether objects of different stiffness levels can be recognized using haptic feedback provided by transcutaneous electrical stimulation of peripheral nerves. Approach. Using a stimulation electrode grid placed along the medial side of the upper arm, the median and ulnar nerve bundles were targeted to evoke haptic sensation on the palmar side of the hand. Stimulation current amplitude was modulated in real-time with the fingertip force recorded from a sensorized prosthetic hand. In order to evaluate which stimulation pattern was more critical, object stiffness was encoded either by the rate of change of the stimulus amplitude or the level of peak stimulus amplitude, as the prosthesis grasped the objects. Main results. Both encoding methods allowed the subjects to differentiate objects of different stiffness levels with  >90% accuracy. No significant difference was observed between the two encoding methods, which indicated that both the rate of change of the stimulation amplitude and the peak stimulation amplitude could effectively provide stiffness information of the objects. Significance. The outcomes suggest that it is possible to elicit haptic sensations describing various object stiffness levels using transcutaneous nerve stimulation. The haptic feedback associated with object stiffness can facilitate object manipulation/interactions. It may also improve user experience during human–machine interactions, when object stiffness information is incorporated.}, number={1}, journal={Journal of Neural Engineering}, publisher={IOP Publishing}, author={Vargas, Luis and Shin, Henry and Huang, He (Helen) and Zhu, Yong and Hu, Xiaogang}, year={2019}, month={Dec}, pages={016002} }
 @inproceedings{li_gao_wen_si_huang_2019, title={Offline Policy Iteration Based Reinforcement Learning Controller for Online Robotic Knee Prosthesis Parameter Tuning}, volume={2019-May}, ISBN={9781538660270}, url={http://dx.doi.org/10.1109/icra.2019.8794212}, DOI={10.1109/icra.2019.8794212}, abstractNote={This paper aims to develop an optimal controller that can automatically provide personalized control of robotic knee prosthesis in order to best support gait of individual prosthesis wearers. We introduced a new reinforcement learning (RL) controller for this purpose based on the promising ability of RL controllers to solve optimal control problems through interactions with the environment without requiring an explicit system model. However, collecting data from a human-prosthesis system is expensive and thus the design of a RL controller has to take into account data and time efficiency. We therefore propose an offline policy iteration based reinforcement learning approach. Our solution is built on the finite state machine (FSM) impedance control framework, which is the most used prosthesis control method in commercial and prototypic robotic prosthesis. Under such a framework, we designed an approximate policy iteration algorithm to devise impedance parameter update rules for 12 prosthesis control parameters in order to meet individual users’ needs. The goal of the reinforcement learning-based control was to reproduce near-normal knee kinematics during gait. We tested the RL controller obtained from offline learning in real time experiment involving the same able-bodied human subject wearing a robotic lower limb prosthesis. Our results showed that the RL control resulted in good convergent behavior in kinematic states, and the offline learning control policy successfully adjusted the prosthesis control parameters to produce near-normal knee kinematics in 10 updates of the impedance control parameters.}, booktitle={2019 International Conference on Robotics and Automation (ICRA)}, publisher={IEEE}, author={Li, Minhan and Gao, Xiang and Wen, Yue and Si, Jennie and Huang, He Helen}, year={2019}, month={May}, pages={2831–2837} }
 @article{wen_si_brandt_gao_huang_2019, title={Online Reinforcement Learning Control for the Personalization of a Robotic Knee Prosthesis}, volume={50}, ISSN={2168-2267 2168-2275}, url={http://dx.doi.org/10.1109/tcyb.2019.2890974}, DOI={10.1109/TCYB.2019.2890974}, abstractNote={Robotic prostheses deliver greater function than passive prostheses, but we face the challenge of tuning a large number of control parameters in order to personalize the device for individual amputee users. This problem is not easily solved by traditional control designs or the latest robotic technology. Reinforcement learning (RL) is naturally appealing. The recent, unprecedented success of AlphaZero demonstrated RL as a feasible, large-scale problem solver. However, the prosthesis-tuning problem is associated with several unaddressed issues such as that it does not have a known and stable model, the continuous states and controls of the problem may result in a curse of dimensionality, and the human-prosthesis system is constantly subject to measurement noise, environmental change and human-body-caused variations. In this paper, we demonstrated the feasibility of direct heuristic dynamic programming, an approximate dynamic programming (ADP) approach, to automatically tune the 12 robotic knee prosthesis parameters to meet individual human users’ needs. We tested the ADP-tuner on two subjects (one able-bodied subject and one amputee subject) walking at a fixed speed on a treadmill. The ADP-tuner learned to reach target gait kinematics in an average of 300 gait cycles or 10 min of walking. We observed improved ADP tuning performance when we transferred a previously learned ADP controller to a new learning session with the same subject. To the best of our knowledge, our approach to personalize robotic prostheses is the first implementation of online ADP learning control to a clinical problem involving human subjects.}, number={6}, journal={IEEE Transactions on Cybernetics}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Wen, Yue and Si, Jennie and Brandt, Andrea and Gao, Xiang and Huang, He}, year={2019}, pages={1–11} }
 @inproceedings{fleming_huang_2019, title={Proportional Myoelectric Control of a Powered Ankle Prosthesis for Postural Control under Expected Perturbation: A Pilot Study}, volume={2019-June}, ISBN={9781728127552}, url={http://dx.doi.org/10.1109/icorr.2019.8779509}, DOI={10.1109/icorr.2019.8779509}, abstractNote={In this study we aimed to investigate the potential for antagonistic residual muscles to generate anticipatory and compensatory postural adjustments and their benefit to postural control with proportional myoelectric control of a powered ankle prosthesis. We conducted this investigation using a predictable pendulum drop task with a single transtibial amputee. In two individual testing sessions the participant used his prescribed passive device and a powered device with pneumatic artificial muscles actuated proportionally to the activation of residual Tibialis Anterior (TA) and Lateral Gastrocnemius (GAS) muscles. Results demonstrated the transtibial amputee generated activations from the residual TA significantly earlier in the powered condition $(p=\theta.\theta\theta 7)$. In the powered condition anticipatory center of pressure excursions were significantly higher $(p=\theta.\theta 17)$, and peak center of mass excursions were reduced $(p=\theta.\theta 21)$. Peak medio-lateral center of pressure excursions were also significantly less in the direction of the intact limb for the powered condition $(p\ =\ \theta.\theta\theta 3)$. The results from this pilot study demonstrate the promise for transtibial amputees to generate anticipatory postural adjustments as well as the potential improvement of stability under expected perturbations. This pilot study provides an initial basis for future study using proportional myoelectric control via antagonistic residual muscles for the control of posture under expected perturbations.}, booktitle={2019 IEEE 16th International Conference on Rehabilitation Robotics (ICORR)}, publisher={IEEE}, author={Fleming, Aaron and Huang, He Helen}, year={2019}, month={Jun}, pages={899–904} }
 @article{fleming_huang_huang_2019, title={Proportional Myoelectric Control of a Virtual Inverted Pendulum Using Residual Antagonistic Muscles: Toward Voluntary Postural Control}, volume={27}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85068905602&partnerID=MN8TOARS}, DOI={10.1109/TNSRE.2019.2922102}, abstractNote={This paper aims to investigate whether transtibial amputees are capable of coordinating the descending neural commands to antagonistic residual ankle muscles for performing dynamic tasks that require continuous, precise control. To achieve this goal, we developed a virtual inverted pendulum that was inherently unstable and mimicked human-like dynamics in a standing posture. Balancing this dynamic system requires continuous inputs, proportional to electromyography (EMG) magnitudes recorded from (residual) tibialis anterior (TA) and lateral gastrocnemius muscles (GAS), respectively. The six able-bodied and six transtibial amputees were recruited and asked to balance the inverted pendulum for ten 90-s trials. The results showed that the amputees were capable of controlling this unstable dynamic system with a proportional myoelectric control; however, they underperformed the able-bodied subjects, who maintained the pendulum closer to center ( ${p} = {0.041}$ ). Compared to the performance in the initial two trials, amputees improved the performance by significantly reducing the number of pendulum falls ( ${p} = {0.0329}$ ) and sway size ( ${p} ={0.048}$ ) in the final two trials. However, the amount of improvement varied across amputee subjects. Amputee subjects demonstrated different task adaptation strategies, including reduction of erroneous residual muscle contractions, development of an appropriate state-action (pendulum state-EMG activation) relationship for the task, and/or reduction of muscle control variability with the improved task performance efficiency (i.e., increased inactivity and sway minimization). The results suggest that after the training of transtibial amputees in coordinating antagonistic residual muscles in dynamic systems, it may be feasible to implement the proportional myoelectric control of the powered ankle prostheses in order to assist the postural control mechanisms, such as anticipatory and compensatory postural adjustments.}, number={7}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Fleming, Aaron and Huang, Stephanie and Huang, He}, year={2019}, month={Jul}, pages={1473–1482} }
 @book{gao_wen_li_si_huang_2019, title={Robotic Knee Parameter Tuning Using Approximate Policy Iteration}, volume={1005}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85065763048&partnerID=MN8TOARS}, DOI={10.1007/978-981-13-7983-3_49}, abstractNote={This paper presents an online model-free reinforcement learning based controller realized by approximate dynamic programming for a robotic knee as part of a human-machine system. Traditionally, prosthesis wearers’ gait performance is improved by manually tuning the impedance parameters. In this paper, we show that the parameter tuning problem can be formulated as an optimal control problem and thus solved by dynamic programming. Toward this goal, we constructed an quadratic instantaneous cost, which resulted in a value function that could be approximated by a neural network. The control policy is then solved by the least-squared method iteratively, a framework of which we refer to as approximate policy iteration. We performed extensive simulations based on prosthetic kinetics and human performance data extracted from real human subjects. Our results show that the proposed parameter tuning algorithm can be readily used for adaptive optimal tuning of prosthetic knee control parameters and the tuning process is time and sample efficient.}, journal={Communications in Computer and Information Science}, author={Gao, X. and Wen, Y. and Li, M. and Si, J. and Huang, H.H.}, year={2019}, pages={554–563} }
 @book{wen_gao_si_brandt_li_huang_2019, title={Robotic Knee Prosthesis Real-Time Control Using Reinforcement Learning with Human in the Loop}, volume={1005}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85065759212&partnerID=MN8TOARS}, DOI={10.1007/978-981-13-7983-3_41}, abstractNote={Advanced robotic prostheses are expensive considering the cost of human resources and the time spent on manually tuning the high-dimensional control parameters for individual users. To alleviate clinicians' effort and promote the advanced robotic prosthesis, we implemented an optimal adaptive control algorithm, which fundamentally is a type of reinforcement learning method, to automatically tune the high-dimensional control parameters of a robotic knee prosthesis through interaction with a human-prosthesis system. The 'human-in-the-loop' term means that the learning controller tunes the control parameters based on the performance of the robotic knee prosthesis while an amputee subject walking with it. We validated the human-in-the-loop auto-tuner with one transfemoral amputee subject for 4 hour-long lab testing sessions. Our results demonstrated that this novel reinforcement learning controller was able to learn through interaction with the human-prosthesis system and discover a set of suitable control parameter for the amputee user to generate near-normative knee kinematics.}, journal={Communications in Computer and Information Science}, author={Wen, Y. and Gao, X. and Si, J. and Brandt, A. and Li, M. and Huang, H.H.}, year={2019}, pages={463–473} }
 @article{huang_huang_2019, title={Voluntary Control of Residual Antagonistic Muscles in Transtibial Amputees: Reciprocal Activation, Coactivation, and Implications for Direct Neural Control of Powered Lower Limb Prostheses}, volume={27}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85058152428&partnerID=MN8TOARS}, DOI={10.1109/TNSRE.2018.2885641}, abstractNote={Residual ankle muscles (i.e., previously antagonistic ankle muscles) of transtibial amputees are a potential source for continuous feedforward control of powered ankle prostheses using proportional myoelectric control. The ability for transtibial amputees to use their residual ankle muscles for two control input degrees of freedom (i.e., two independent myoelectric control input sources) for direct neural control depends on the ability for amputees to generate varying magnitudes of reciprocal activation and coactivation using their residual ankle muscles, which is not well understood. In this paper, we aimed to fill this knowledge gap. We asked 12 transtibial amputees to control the 2-D movement of a computer cursor using continuous proportional myoelectric control via their residual plantar flexor and residual dorsiflexor muscles to define their reachable 2-D control input space. The x–y position of the computer cursor was directly proportional to the independent continuous myoelectric control signals from the residual lateral gastrocnemius (x-axis) and the residual tibialis anterior (y-axis) where the limits of each axis were 0%–100% maximum voluntary activation of the corresponding residual muscle. Our results show that the reachable control input space varied widely across amputee subjects ranging from 38% to 81% of the maximum possible control input space. The cumulative time for the amputee subjects to saturate their reachable control input space ranged from 1.95 to 6.85 min. The amputee subjects used different residual muscle activation patterns and coordination strategies to expand their reachable control input space depending on their ability to perform coactivation and reciprocal activation using their residual plantar flexor and dorsiflexor muscles. The future development of powered lower limb prostheses using direct continuous proportional myoelectric control via residual muscles (e.g., for direct voluntary control of prosthesis joint impedance) should consider how an amputee user’s immediately accessible residual muscle activation patterns and reachable 2-D control input space may affect their learning and performance.}, number={1}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Huang, Stephanie and Huang, He}, year={2019}, month={Jan}, pages={85–95} }
 @article{zhang_huang_2018, title={A Lower-Back Robotic Exoskeleton}, volume={25}, ISSN={["1558-223X"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85047189327&partnerID=MN8TOARS}, DOI={10.1109/mra.2018.2815083}, abstractNote={A lower-back exoskeleton prototype designed to provide back support for industrial workers who manually handle heavy materials is presented in this article. Reducing spinal loads during these tasks can reduce the risk of work-related back injuries. Biomechanical studies show that compression of the lumbar spine is a key risk factor for musculoskeletal injuries. To address this issue, we present a wearable exoskeleton designed to provide back support and reduce lumbar spine compression. To provide effective assistance and avoid injury to muscles or tendons, we apply a continuous torque of approximately 40 Nm on both hip joints to actively assist both hip abduction/adduction (HAA) and hip flexion/extension (HFE). Each actuation unit includes a modular and a compact series-elastic actuator (SEA) with a clutch. The SEA provides mechanical compliance at the interface between the exoskeleton and the user, and the clutches can automatically disengage the torque between the exoskeleton and the user. These experimental results show that the exoskeleton can lower lumbar compression by reducing the need for muscular activity in the spine. Furthermore, powering both HFE and HAA can effectively reduce the lumbar spinal loading user experience when lifting and lowering objects while in a twisted posture.}, number={2}, journal={IEEE ROBOTICS & AUTOMATION MAGAZINE}, author={Zhang, Ting and Huang, He}, year={2018}, month={Jun}, pages={95–106} }
 @inproceedings{vempala_liu_kamper_huang_2018, title={A Practical Approach for Evaluation of Socket Pistoning for Lower Limb Amputees}, volume={2018-July}, ISBN={9781538636466}, url={http://dx.doi.org/10.1109/embc.2018.8513249}, DOI={10.1109/embc.2018.8513249}, abstractNote={Although fit of the socket-suspension system is critical to lower limb amputees, monitoring of the fit is largely based on user feedback, which is subjective and often unreliable. Pistoning, defined as the relative displacement between the socket and the residual limb, is a well-accepted indicator of the fit of the socket-suspension system. However, opacity and rigidity of everyday prosthetic sockets make measurement of pistoning a challenging task. In this paper, we describe the development of a pistoning evaluation procedure that relies on two motion capture systems: a magnetic motion capture system used to measure the motion of the residual limb and an optical motion capture system used to measure the motion of the socket. Through synchronization of the two motion capture systems, the motion of the residual limb relative to the socket can be determined to derive the amplitude of pistoning. Here, we evaluated the performance of our approach through repeated calibration and a treadmill walking task with an amputee. Results demonstrate that this procedure, which does not rely on radiography unlike some existing methods, can be used to evaluate the fit of amputees' everyday sockets.}, booktitle={2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Vempala, Vibhavari and Liu, Ming and Kamper, Derek and Huang, He}, year={2018}, month={Jul}, pages={3938–3941} }
 @inproceedings{pan_harmody_huang_2018, title={A Reliable Multi-User EMG Interface Based on A Generic-Musculoskeletal Model against Loading Weight Changes *}, volume={2018-July}, ISBN={9781538636466}, url={http://dx.doi.org/10.1109/embc.2018.8512685}, DOI={10.1109/embc.2018.8512685}, abstractNote={The reliability of myoelectric control is important to ensure the performance of prostheses during daily use. Recently, we proposed a multi-user neural-machine interface based on a generic musculoskeletal model to simultaneously and continuously estimate flexion/extension movements at the metacarpophalangeal (MCP) and wrist joints from surface electromyography (EMG) signals. Our previous results demonstrated that the multi-user EMG interface was reliable against upper limb posture changes. However, the reliability of the interface against different loading weights, which is an important factor that would decrease the performance of myoelectric control and be tested during the occupational therapy for myoelectric prosthesis users, is still unclear. In this study, we aimed to evaluate the reliability of the generic model over different loading weights. Four able-bodied subjects were tested in this study. Subjects performed a virtual hand/wrist posture matching task with three different loading weights (no weight, 1.25 Lbs, and 2.5 Lbs). All subjects accomplished all the assigned virtual tasks. The on-line experimental results showed that performance with different loading weights was very close. The results demonstrated that the multi-user EMG interface was reliable against the different loading weights, indicating it has potential to promote the myoelectric control into clinical applications.}, booktitle={2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Pan, Lizhi and Harmody, Andrew and Huang, He}, year={2018}, month={Jul}, pages={2104–2107} }
 @inproceedings{fylstra_dai_hu_huang_2018, title={Characterizing Residual Muscle Properties in Lower Limb Amputees Using High Density EMG Decomposition: A Pilot Study*}, volume={2018-January}, ISBN={9781538636466}, url={http://dx.doi.org/10.1109/embc.2018.8513661}, DOI={10.1109/embc.2018.8513661}, abstractNote={As research is progressing towards EMG control of lower limb prostheses, it is vital to understand the neurophysiology of the residual muscles in the amputated limb, which has been largely ignored. Therefore, the goal of this study was to characterize the activation patterns (muscle recruitment and motor unit discharge patterns) of the residual muscles of lower limb amputees. One transtibial amputee subject was recruited for this pilot study. The participant wore three high-density EMG electrode pads (8x8 grid with 64 channels) on each limb (a total of six pads) – one on the tibialis anterior (TA), medial gastrocnemius (MG), and lateral gastrocnemius (LG), respectively. The participant was asked to follow a ramping procedure plateauing at 50% of maximum voluntary contraction (MVC) for both the TA and Gastrocnemius muscles. The EMG signals were then decomposed offline; the firing rate and spatial activation patterns of the muscle were analyzed. Results showed slower and more variable firing rate in motor units of residual muscles than those of intact side. In addition, the spatial pattern of muscle activation differed between residual and intact muscles. These results indicate that surface EMG signals recorded from residual muscles present modified signal features from intact shank muscles, which should be considered when implementing myoelectric control schemes.}, booktitle={2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Fylstra, Bretta L. and Dai, Chenyun and Hu, Xiaogang and Huang, He Helen}, year={2018}, month={Jul}, pages={5974–5977} }
 @article{crouch_pan_filer_stallings_huang_2018, title={Comparing Surface and Intramuscular Electromyography for Simultaneous and Proportional Control Based on a Musculoskeletal Model: A Pilot Study}, volume={26}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85050633647&partnerID=MN8TOARS}, DOI={10.1109/TNSRE.2018.2859833}, abstractNote={Simultaneous and proportional control (SPC) of neural-machine interfaces uses magnitudes of smoothed electromyograms (EMG) as control inputs. Though surface EMG (sEMG) electrodes are common for clinical neural-machine interfaces, intramuscular EMG (iEMG) electrodes may be indicated in some circumstances (e.g., for controlling many degrees of freedom). However, differences in signal characteristics between sEMG and iEMG may influence SPC performance. We conducted a pilot study to determine the effect of electrode type (sEMG and iEMG) on real-time task performance with SPC based on a novel 2-degree-of-freedom EMG-driven musculoskeletal model of the wrist and hand. Four able-bodied subjects and one transradial amputee performed a virtual posture matching task with either sEMG or iEMG. There was a trend of better task performance with sEMG than iEMG for both able-bodied and amputee subjects, though the difference was not statistically significant. Thus, while iEMG may permit targeted recording of EMG, its signal characteristics may not be as ideal for SPC as those of sEMG. The tradeoff between recording specificity and signal characteristics is an important consideration for development and clinical implementation of SPC for neural-machine interfaces.}, number={9}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Crouch, Dustin L. and Pan, Lizhi and Filer, William and Stallings, Jonathan W. and Huang, He}, year={2018}, month={Sep}, pages={1735–1744} }
 @inproceedings{fleming_huang_huang_2018, title={Coordination of Voluntary Residual Muscle Contractions in Transtibial Amputees: a Pilot Study}, volume={2018-July}, ISBN={9781538636466}, url={http://dx.doi.org/10.1109/embc.2018.8512674}, DOI={10.1109/embc.2018.8512674}, abstractNote={Recently there has been considerable interest in the use of electromyography (EMG) for the control of powered, lower-limb prostheses. However, little is understood regarding amputee residual muscle, specifically the ability for lower-limb amputees to coordinate previously antagonist residual muscles for different control tasks. In this study, we aimed to investigate the capability of transtibial amputees in coordinating residual gastrocnemius and tibialis anterior for performing a high-level task. Specifically, we examined how three transtibial amputees and one ablebodied subject used residual and intact ankle muscles to balance a virtual inverted pendulum. Subjects controlled the pendulum by modulating stiffness of the base joint proportional to the level of EMG signal exerted. We conducted ten trials for each subject and quantified success of task performance by area of overall sway and number of falls (termed failure). Amputees successfully reduced number of failures, though not to the extent of able-bodied subjects. Interestingly, while able-bodied subjects reduced overall pendulum sway, amputees did not. EMG coordination patterns that preceded failure for amputees were different than that of able-bodied subjects. These results suggest that amputees have altered ability to coordinate muscle post-amputation; however, all subjects can improve the task performance and learn to reduce EMG coordination patterns that led to task failure. Further study is required to investigate the limit of amputees in learning of coordination between antagonist residual muscles in order to inform future neural control of prosthetic legs}, booktitle={2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Fleming, Aaron and Huang, Stephanie and Huang, He}, year={2018}, month={Jul}, pages={2128–2131} }
 @article{zhang_tran_huang_2018, title={Design and Experimental Verification of Hip Exoskeleton With Balance Capacities for Walking Assistance}, volume={23}, ISSN={["1941-014X"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85040563606&partnerID=MN8TOARS}, DOI={10.1109/tmech.2018.2790358}, abstractNote={Most current hip exoskeletons emphasize assistance for walking rather than stability. The goal of this paper is to develop a novel, high-power, self-balancing, and passively and software-controlled actively compliant hip exoskeleton that can assist with movement and maintain balance in both the sagittal and frontal planes. The developed hip exoskeleton includes powered hip abduction/adduction and hip flexion/extension joints. Each actuation unit employs a modular and compact series elastic actuator (SEA) with a high torque-to-weight ratio. It provides mechanical compliance at the interface between the exoskeleton and the wearer to ensure safety and a natural gait in the coupled wearer-exoskeleton system. A new balance controller based on the extrapolated center of mass concept is presented for maintaining walking stability. This controller reacts to perturbations in balance and produces a compliant guidance force through a combination of the passive elasticity of the SEA and active compliant control based on adaptive admittance control. The function of the hip exoskeleton is not to override human control, but rather to involve the wearer in movement control in order to avoid conflicts between wearer and exoskeleton. Our preliminary experiments on a healthy subject wearing the hip exoskeleton demonstrate the potential effectiveness of the proposed hip exoskeleton and controller for walking balance control.}, number={1}, journal={IEEE-ASME TRANSACTIONS ON MECHATRONICS}, author={Zhang, Ting and Tran, Minh and Huang, He}, year={2018}, month={Feb}, pages={274–285} }
 @article{resnik_huang_winslow_crouch_zhang_wolk_2018, title={Evaluation of EMG pattern recognition for upper limb prosthesis control: a case study in comparison with direct myoelectric control}, volume={15}, ISSN={["1743-0003"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85043758461&partnerID=MN8TOARS}, DOI={10.1186/s12984-018-0361-3}, abstractNote={Although electromyogram (EMG) pattern recognition (PR) for multifunctional upper limb prosthesis control has been reported for decades, the clinical benefits have rarely been examined. The study purposes were to: 1) compare self-report and performance outcomes of a transradial amputee immediately after training and one week after training of direct myoelectric control and EMG pattern recognition (PR) for a two-degree-of-freedom (DOF) prosthesis, and 2) examine the change in outcomes one week after pattern recognition training and the rate of skill acquisition in two subjects with transradial amputations.In this cross-over study, participants were randomized to receive either PR control or direct control (DC) training of a 2 DOF myoelectric prosthesis first. Participants were 2 persons with traumatic transradial (TR) amputations who were 1 DOF myoelectric users. Outcomes, including measures of dexterity with and without cognitive load, activity performance, self-reported function, and prosthetic satisfaction were administered immediately and 1 week after training. Speed of skill acquisition was assessed hourly. One subject completed training under both PR control and DC conditions. Both subjects completed PR training and testing. Outcomes of test metrics were analyzed descriptively.Comparison of the two control strategies in one subject who completed training in both conditions showed better scores in 2 (18%) dexterity measures, 1 (50%) dexterity measure with cognitive load, and 1 (50%) self-report functional measure using DC, as compared to PR. Scores of all other metrics were comparable. Both subjects showed decline in dexterity after training. Findings related to rate of skill acquisition varied considerably by subject.Outcomes of PR and DC for operating a 2-DOF prosthesis in a single subject cross-over study were similar for 74% of metrics, and favored DC in 26% of metrics. The two subjects who completed PR training showed decline in dexterity one week after training ended. Findings related to rate of skill acquisition varied considerably by subject. This study, despite its small sample size, highlights a need for additional research quantifying the functional and clinical benefits of PR control for upper limb prostheses.}, number={1}, journal={JOURNAL OF NEUROENGINEERING AND REHABILITATION}, author={Resnik, Linda and Huang, He and Winslow, Anna and Crouch, Dustin L. and Zhang, Fan and Wolk, Nancy}, year={2018}, month={Mar} }
 @article{shin_watkins_huang_zhu_hu_2018, title={Evoked haptic sensations in the hand via non-invasive proximal nerve stimulation}, volume={15}, ISSN={["1741-2552"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85049836114&partnerID=MN8TOARS}, DOI={10.1088/1741-2552/aabd5d}, abstractNote={Objective. Haptic perception of a prosthetic limb or hand is a crucial, but often unmet, need which impacts the utility of the prostheses. In this study, we seek to evaluate the feasibility of a non-invasive transcutaneous nerve stimulation method in generating haptic feedback in a transradial amputee subject as well as intact able-bodied subjects. Approach. An electrode grid was placed on the skin along the medial side of the upper arm beneath the short head of the biceps brachii, in proximity to the median and ulnar nerves. Varying stimulation patterns were delivered to different electrode pairs, in order to emulate different types of sensations (Single Tap, Press-and-Hold, Double Tap) at different regions of the hand. Subjects then reported the magnitude of sensation by pressing on a force transducer to transform the qualitative haptic perception into a quantitative measurement. Main results. Altering current stimulations through electrode pairs on the grid resulted in repeatable alterations in the percept regions of the hand. Most subjects reported spatial coverage of individual fingers or phalanges, which can resemble the whole hand through different pairs of stimulation electrodes. The different stimulation patterns were also differentiable by all subjects. The amputee subject also reported haptic sensations similar to the able-bodied subjects. Significance. Our findings demonstrated the capabilities of our transcutaneous stimulation method. Subjects were able to perceive spatially distinct sensations with graded magnitudes that emulated tapping and holding sensation in their hands. The elicitation of haptic sensations in the phantom hand of an amputee is a significant step in the development of our stimulation method, and provides insight into the future adaptation and implementation of prostheses with non-invasive sensory feedback to the users.}, number={4}, journal={JOURNAL OF NEURAL ENGINEERING}, author={Shin, Henry and Watkins, Zach and Huang, He and Zhu, Yong and Hu, Xiaogang}, year={2018}, month={Aug} }
 @article{islam_stallings_staicu_crouch_pan_huang_2018, title={Functional variable selection for emg-based control of a robotic hand prosthetic}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85095037692&partnerID=MN8TOARS}, journal={arXiv}, author={Islam, M.N. and Stallings, J. and Staicu, A.-M. and Crouch, D. and Pan, L. and Huang, H.}, year={2018} }
 @inproceedings{vargas_huang_zhu_hu_2018, title={Merged Haptic Sensation in the Hand during Concurrent Non-Invasive Proximal Nerve Stimulation}, volume={2018-July}, ISBN={9781538636466}, url={http://dx.doi.org/10.1109/embc.2018.8512707}, DOI={10.1109/embc.2018.8512707}, abstractNote={When individuals interact with the environment, sensory feedback is a critical aspect of the experience. Individuals using prosthesis often have difficulty controlling their device, partly due to a lack of sensory information. Transcutaneous nerve stimulation has the potential to elicit focal haptic sensation when controlled electrical current was delivered to a pair of electrodes in proximity to the nerve. The objective of this preliminary study was to evaluate how different elicited focal haptic sensation were altered, when multiple concurrent electrical stimuli were delivered to different portions of the median and ulnar nerve bundles. The delay between the individual stimulation during concurrent stimuli was also varied to identify if this parameter could alter the resulting sensation region. Lastly, the stability/repeatability of the perceived sensation during concurrent stimuli was determined. Our preliminary results showed that the spatial distribution of the haptic sensation was largely a direct summation/merge of the sensation regions from the individual nerve stimulation when comparing the regions to that of the concurrent double stimulation. Our results also showed that merged sensation region was not sensitive to different time delays the two concurrent stimuli. Lastly, the sensation regions remained stable and showed repeatable sensation in the hand even with 20–60 minutes between repeated stimulations.}, booktitle={2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Vargas, Luis and Huang, He Helen and Zhu, Yong and Hu, Xiaogang}, year={2018}, month={Jul}, pages={2186–2189} }
 @article{pan_crouch_huang_2018, title={Myoelectric Control Based on a Generic Musculoskeletal Model: Toward a Multi-User Neural-Machine Interface}, volume={26}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85047202264&partnerID=MN8TOARS}, DOI={10.1109/tnsre.2018.2838448}, abstractNote={This paper aimed to develop a novel electromyography (EMG)-based neural-machine interface (NMI) that is user-generic for continuously predicting coordinated motion betweenmuscle contractionmetacarpophalangeal (MCP) and wrist flexion/extension. The NMI requires a minimum calibration procedure that only involves capturing maximal voluntary muscle contraction for themonitoredmuscles for individual users. At the center of the NMI is a user-generic musculoskeletal model based on the experimental data collected from six able-bodied (AB) subjects and nine different upper limb postures. The generic model was evaluated on-line on both AB subjects and a transradial amputee. The subjectswere instructed to performa virtual hand/wrist posture matching task with different upper limb postures. The on-line performanceof the genericmodelwas also compared with that of the musculoskeletal model customized to each individual user (called “specific model”). All subjects accomplished the assigned virtual tasks while using the user-generic NMI, although the AB subjects produced better performance than the amputee subject. Interestingly, compared with the specific model, the generic model produced comparable completion time, a reduced number of overshoots, and improved path efficiency in the virtual hand/wrist posture matching task. The results suggested that it is possible to design an EMG-driven NMI based on a musculoskeletalmodelthat could fit multiple users, including upper limb amputees, for predicting coordinated MCP and wrist motion. The present new method might address the challenges of existing advanced EMG-based NMI that require frequent and lengthy customization and calibration. Our future research will focus on evaluating the developed NMI for powered prosthetic arms.}, number={7}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Pan, Lizhi and Crouch, Dustin L. and Huang, He}, year={2018}, month={Jul}, pages={1435–1442} }
 @inproceedings{diaz_da silva_zhong_huang_lobaton_2018, title={Visual Terrain Identification and Surface Inclination Estimation for Improving Human Locomotion with a Lower-Limb Prosthetic}, volume={2018-July}, ISBN={9781538636466}, url={http://dx.doi.org/10.1109/embc.2018.8512614}, DOI={10.1109/embc.2018.8512614}, abstractNote={Lower-limb robotic prosthetics can benefit from context awareness to provide comfort and safety to the amputee. In this work, we developed a terrain identification and surface inclination estimation system for a prosthetic leg using visual and inertial sensors. We built a dataset from which images with high sharpness are selected using the IMU signal. The images are used for terrain identification and inclination is also computed simultaneously. With such information, the control of a robotized prosthetic leg can be adapted to changes in its surrounding.}, booktitle={2018 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Diaz, Jean P. and da Silva, Rafael L. and Zhong, Boxuan and Huang, He Helen and Lobaton, Edgar}, year={2018}, month={Jul}, pages={1817–1820} }
 @article{huang_huang_2018, title={Voluntary Control of Residual Antagonistic Muscles in Transtibial Amputees: Feedforward Ballistic Contractions and Implications for Direct Neural Control of Powered Lower Limb Prostheses}, volume={26}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85042875741&partnerID=MN8TOARS}, DOI={10.1109/tnsre.2018.2811544}, abstractNote={Discrete, rapid (i.e., ballistic like) muscle activation patterns have been observed in ankle muscles (i.e., plantar flexors and dorsiflexors) of able-bodied individuals during voluntary posture control. This observation motivated us to investigate whether transtibial amputees are capable of generating such a ballistic-like activation pattern accurately using their residual ankle muscles in order to assess whether the volitional postural control of a powered ankle prosthesis using proportional myoelectric control via residual muscles could be feasible. In this paper, we asked ten transtibial amputees to generate ballistic-like activation patterns using their residual lateral gastrocnemius and residual tibialis anterior to control a computer cursor via proportional myoelectric control to hit targets positioned at 20% and 40% of maximum voluntary contraction of the corresponding residual muscle. During practice conditions, we asked amputees to hit a single target repeatedly. During testing conditions, we asked amputees to hit a random sequence of targets. We compared movement time to target and end-point accuracy. We also examined motor recruitment synchronization via time-frequency representations of residual muscle activation. The result showed that median end-point error ranged from −0.6% to 1% maximum voluntary contraction across subjects during practice, which was significantly lower compared to testing ( $p < 0.001$ ). Average movement time for all amputees was 242 ms during practice and 272 ms during testing. Motor recruitment synchronization varied across subjects, and amputees with the highest synchronization achieved the fastest movement times. End-point accuracy was independent of movement time. Results suggest that it is feasible for transtibial amputees to generate ballistic control signals using their residual muscles. Future work on volitional control of powered power ankle prostheses might consider anticipatory postural control based on ballistic-like residual muscle activation patterns and direct continuous proportional myoelectric control.}, number={4}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Huang, Stephanie and Huang, He}, year={2018}, month={Apr}, pages={894–903} }
 @article{wen_si_gao_huang_huang_2017, title={A New Powered Lower Limb Prosthesis Control Framework Based on Adaptive Dynamic Programming}, volume={28}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84978229791&partnerID=MN8TOARS}, DOI={10.1109/TNNLS.2016.2584559}, abstractNote={This brief presents a novel application of adaptive dynamic programming (ADP) for optimal adaptive control of powered lower limb prostheses, a type of wearable robots to assist the motor function of the limb amputees. Current control of these robotic devices typically relies on finite state impedance control (FS-IC), which lacks adaptability to the user’s physical condition. As a result, joint impedance settings are often customized manually and heuristically in clinics, which greatly hinder the wide use of these advanced medical devices. This simulation study aimed at demonstrating the feasibility of ADP for automatic tuning of the twelve knee joint impedance parameters during a complete gait cycle to achieve balanced walking. Given that the accurate models of human walking dynamics are difficult to obtain, the model-free ADP control algorithms were considered. First, direct heuristic dynamic programming (dHDP) was applied to the control problem, and its performance was evaluated on OpenSim, an often-used dynamic walking simulator. For the comparison purposes, we selected another established ADP algorithm, the neural fitted Q with continuous action (NFQCA). In both cases, the ADP controllers learned to control the right knee joint and achieved balanced walking, but dHDP outperformed NFQCA in this application during a 200 gait cycle-based testing.}, number={9}, journal={IEEE Transactions on Neural Networks and Learning Systems}, author={Wen, Y. and Si, J. and Gao, X. and Huang, S. and Huang, H.H.}, year={2017}, pages={2215–2220} }
 @article{wen_si_gao_huang_huang_2017, title={A new powered lower limb prosthesis control framework based on adaptive dynamic programming}, volume={28}, number={9}, journal={IEEE Transactions on Neural Networks and Learning Systems}, author={Wen, Y. and Si, J. and Gao, X. and Huang, S. and Huang, H.}, year={2017}, pages={2215–2220} }
 @inproceedings{agcayazi_mcknight_sotory_huang_ghosh_bozkurt_2017, title={A scalable shear and normal force sensor for prosthetic sensing}, volume={2017-December}, ISBN={9781509010127}, url={http://dx.doi.org/10.1109/icsens.2017.8233977}, DOI={10.1109/icsens.2017.8233977}, abstractNote={Current techniques of stress measurement to assess comfort for prosthetic limbs focus on resolving different forces as well as the directions for each force. The sophistication introduced by spatially scaling these techniques for the entire interface with the difficulty of integration in the current prosthetic limbs call for a more intelligent design. In this work, we introduce the design and experimental results of a capacitive shear and normal force sensor. Our force sensor works with a single transduction point to ease the process of spatial scaling and is integrated seamlessly on the liner component of the prosthetic limb. We provide proof-of-concept results of both shear and normal force sensing with our capacitive force sensor.}, booktitle={2017 IEEE SENSORS}, publisher={IEEE}, author={Agcayazi, Talha and McKnight, Michael and Sotory, Peter and Huang, Helen and Ghosh, Tushar and Bozkurt, Alper}, year={2017}, month={Oct}, pages={1–3} }
 @article{liu_zhang_huang_2017, title={An Adaptive Classification Strategy for Reliable Locomotion Mode Recognition}, volume={17}, ISSN={1424-8220}, url={http://dx.doi.org/10.3390/s17092020}, DOI={10.3390/s17092020}, abstractNote={Algorithms for locomotion mode recognition (LMR) based on surface electromyography and mechanical sensors have recently been developed and could be used for the neural control of powered prosthetic legs. However, the variations in input signals, caused by physical changes at the sensor interface and human physiological changes, may threaten the reliability of these algorithms. This study aimed to investigate the effectiveness of applying adaptive pattern classifiers for LMR. Three adaptive classifiers, i.e., entropy-based adaptation (EBA), LearnIng From Testing data (LIFT), and Transductive Support Vector Machine (TSVM), were compared and offline evaluated using data collected from two able-bodied subjects and one transfemoral amputee. The offline analysis indicated that the adaptive classifier could effectively maintain or restore the performance of the LMR algorithm when gradual signal variations occurred. EBA and LIFT were recommended because of their better performance and higher computational efficiency. Finally, the EBA was implemented for real-time human-in-the-loop prosthesis control. The online evaluation showed that the applied EBA effectively adapted to changes in input signals across sessions and yielded more reliable prosthesis control over time, compared with the LMR without adaptation. The developed novel adaptive strategy may further enhance the reliability of neurally-controlled prosthetic legs.}, number={9}, journal={Sensors}, publisher={MDPI AG}, author={Liu, M. and Zhang, F. and Huang, H.}, year={2017}, month={Sep}, pages={2020} }
 @inproceedings{wen_brandt_si_huang_2018, title={Automatically customizing a powered knee prosthesis with human in the loop using adaptive dynamic programming}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85049991814&partnerID=MN8TOARS}, DOI={10.1109/WEROB.2017.8383835}, abstractNote={In this study, we validated a human-in-the-loop auto-tuner using machine learning to automatically customize powered knee prosthesis control parameters for an amputee subject in real time. The experimental powered knee prosthesis was controlled by a finite state impedance controller, which had 12 configurable impedance control parameters. Using adaptive dynamic programming with reinforcement learning while one transfemoral amputee subject walked with the powered knee prosthesis, the auto-tuner would interact with the human-prosthesis system and learn to configure the high dimension of control parameters. We tested 4 different initial conditions with one unilateral transfemoral amputee subject. The results showed that the auto-tuner discovered the control parameters that allowed amputee subject to generate normative knee kinematics in 3 out of 4 tuning sessions. For all test sessions, the averaged root-mean-square error of the knee kinematics relative to the normative knee kinematics decreased from 6.6 degrees to 4.6 degrees after tuning procedure.}, booktitle={2017 International Symposium on Wearable Robotics and Rehabilitation, WeRob 2017}, author={Wen, Y. and Brandt, A. and Si, J. and Huang, H.H.}, year={2018}, pages={1–2} }
 @inproceedings{wen_brandt_liu_huang_si_2017, title={Comparing parallel and sequential control parameter tuning for a powered knee prosthesis}, volume={2017-January}, ISBN={9781538616451}, url={http://dx.doi.org/10.1109/smc.2017.8122863}, DOI={10.1109/smc.2017.8122863}, abstractNote={Powered knee prostheses, compared to traditional energetically-passive knee prostheses, greatly enhance the mobility of transfemoral amputees. However, powered prostheses have a large number of control parameters that must be adjusted for individual amputee users, which presents a great challenge for clinical use. To address this challenge, we proposed and compared 2 automatic tuning strategies (i.e. parallel and sequential) using our newly developed optimal adaptive dynamic programming (ADP) tuner that objectively tuned the control parameters of an experimental powered knee prosthesis to mimic the knee profile of an able-bodied person (i.e. reference profile). With the parallel tuning strategy, we tuned all control parameters during the stance and the swing phases simultaneously. With the sequential tuning strategy, we alternately tuned stance or swing phase control parameters while fixing the remaining parameters. One able-bodied subject with a prosthesis adapter and one transfemoral amputee subject walked with the experimental powered knee prosthesis under both tuning strategies. Results show that with both tuning strategies, the ADP tuner successfully tuned the impedance parameters to match the prosthetic knee profile to the reference profile. Additionally, the parallel strategy outperformed the sequential strategy with better convergence to the reference profile. Interestingly, with the sequential tuning strategy, tuning during the swing phase greatly impacted the subsequent stance phase profile, but the impact was not as great when the order of tuning was switched. The ability to simultaneously adjust all control parameters with ADP using a parallel strategy may be a preferred solution for the current high-dimension control challenge, which may lead to more advanced, adaptive powered knee prostheses.}, booktitle={2017 IEEE International Conference on Systems, Man, and Cybernetics (SMC)}, publisher={IEEE}, author={Wen, Yue and Brandt, Andrea and Liu, Ming and Huang, He Helen and Si, Jennie}, year={2017}, month={Oct}, pages={1716–1721} }
 @inproceedings{chung_crouch_huang_2017, title={Effects of output speed threshold on real-time continuous EMG human-machine interface control}, volume={2017-January}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85044244968&partnerID=MN8TOARS}, DOI={10.1109/smc.2017.8122805}, abstractNote={Continuous EMG control of human-machine interfaces (HMIs) enables more direct and flexible control of output movements than discrete classification algorithms. However, EMG is a non-stationary signal and can add noise to continuous EMG control output. We studied the effect of an output speed threshold method to stabilize the movement prediction of a 2-DOF musculoskeletal model-based continuous EMG controller during a real-time virtual task. In each of several trials, three able-bodied subjects were instructed to move and align the palm and finger segments of a virtual hand with four different target postures on a computer screen. Three different thresholds on the model's predicted angular speed were applied in a randomized order across trials: no threshold, medium threshold (15 °/sec), and high threshold (30 °/sec); the virtual hand did not move if the predicted angular speed at the next timepoint did not exceed the threshold. We recorded completion time, overshoot, jerk, and number of failed trials to quantify task performance. In a separate block of trials, subjects reported their threshold preference following multiple pairwise comparisons. The number of overshoots decreased and jerk magnitude increased with higher threshold levels. The average completion time was lowest with the medium threshold for 2 subjects. All 3 subjects had lower failed trials with either the medium or high threshold. The subject preference score showed an inverse trend with the number of failed trials. In summary, the presented threshold method was successful in reducing overshoot and trial failures, and a threshold was preferred by all subjects over no threshold. Thus, an output speed threshold may improve the functional performance and user satisfaction of continuous EMG control for HMIs, such as powered upper limb prostheses.}, booktitle={Ieee international conference on systems man and cybernetics conference}, author={Chung, S. H. and Crouch, D. L. and Huang, He}, year={2017}, pages={1375–1380} }
 @inproceedings{zhang_huang_2018, title={Enhancing gait balance via a 4-DoFs wearable hip exoskeleton}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85049976192&partnerID=MN8TOARS}, DOI={10.1109/WEROB.2017.8383824}, abstractNote={One limitation of the current lower-limb exoskeletons is that they do not provide the function of maintaining the lateral stability. During walking, beyond the forward step length regulated by hip flexion/extension (HFE), adaptation of the step width, which can be adjusted by hip abduction/adduction (HAA) motions, is also crucial for walking stability. Biomechanical studies have indicated that the step width and the mediolateral foot placement at the end of each step can be estimated based on the center of mass (CoM), which is assumed to be located at the pelvis. The extrapolated center of mass (XCoM) is obtained by vertically projecting the CoM's position to the ground in the direction of its velocity. The present study is to develop a novel, high-power, self-balancing, and passively and software-controlled actively compliant hip exoskeleton that can assist with movement and maintain balance in both the sagittal and frontal planes.}, booktitle={2017 International Symposium on Wearable Robotics and Rehabilitation, WeRob 2017}, author={Zhang, T. and Huang, He}, year={2018}, pages={1–2} }
 @article{brandt_wen_liu_stallings_huang_2017, title={Interactions Between Transfemoral Amputees and a Powered Knee Prosthesis During Load Carriage}, volume={7}, ISSN={2045-2322}, url={http://dx.doi.org/10.1038/S41598-017-14834-7}, DOI={10.1038/S41598-017-14834-7}, abstractNote={AbstractMachines and humans become mechanically coupled when lower limb amputees walk with powered prostheses, but these two control systems differ in adaptability. We know little about how they interact when faced with real-world physical demands (e.g. carrying loads). Here, we investigated how each system (i.e. amputee and powered prosthesis) responds to changes in the prosthesis mechanics and gravitational load. Five transfemoral amputees walked with and without load (i.e. weighted backpack) and a powered knee prosthesis with two pre-programmed controller settings (i.e. for load and no load). We recorded subjects’ kinematics, kinetics, and perceived exertion. Compared to the no load setting, the load setting reduced subjects’ perceived exertion and intact-limb stance time when they carried load. When subjects did not carry load, their perceived exertion and gait performance did not significantly change with controller settings. Our results suggest transfemoral amputees could benefit from load-adaptive powered knee controllers, and controller adjustments affect amputees more when they walk with (versus without) load. Further understanding of the interaction between powered prostheses, amputee users, and various environments may allow researchers to expand the utility of prostheses beyond simple environments (e.g. firm level ground without load) that represent only a subset of real-world environments.}, number={1}, journal={Scientific Reports}, publisher={Springer Nature}, author={Brandt, Andrea and Wen, Yue and Liu, Ming and Stallings, Jonathan and Huang, He Helen}, year={2017}, month={Nov} }
 @inproceedings{pan_crouch_huang_2017, title={Musculoskeletal model for simultaneous and proportional control of 3-DOF hand and wrist movements from EMG signals}, ISBN={9781509046034}, url={http://dx.doi.org/10.1109/ner.2017.8008356}, DOI={10.1109/ner.2017.8008356}, abstractNote={Recently, we proposed a musculoskeletal model to simultaneously predict motion along metacarpophalangeal (MCP) and wrist flexion/extension degrees-of-freedom (DOFs) from surface electromyography (EMG) signals. Since wrist pronation/supination is also functionally important, we extended the musculoskeletal model to simultaneously estimate wrist pronation/supination in addition to wrist and MCP flexion/extension from surface EMG signals of six corresponding muscles. Kinematic data and surface EMG signals were acquired synchronously from an able-bodied subject. The subject was instructed to perform single-DOF movements at fixed or variable speed and simultaneous 3-DOF movements at variable speed during the experiment. The model included six Hill-type actuators, each with a contractile element and a parallel elastic element. Seven parameters were optimized for each of the six muscles. The average Pearson's correlation coefficient (r) between measured and estimated joint angles across all trials was 0.91, indicating high positive correlation. The results demonstrated that the proposed model could feasibly simultaneously estimate 3-DOF joint angles during either independent-DOF or simultaneous 3-DOF movements from EMG signals. Our results promote the potential of the EMG-driven musculoskeletal model for clinical applications, such as prosthesis control.}, booktitle={2017 8th International IEEE/EMBS Conference on Neural Engineering (NER)}, publisher={IEEE}, author={Pan, Lizhi and Crouch, Dustin and Huang, He}, year={2017}, month={May}, pages={325–328} }
 @article{crouch_huang_2017, title={Musculoskeletal model-based control interface mimics physiologic hand dynamics during path tracing task}, volume={14}, ISSN={["1741-2552"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85020447492&partnerID=MN8TOARS}, DOI={10.1088/1741-2552/aa61bc}, abstractNote={Objective. We investigated the feasibility of a novel, customizable, simplified EMG-driven musculoskeletal model for estimating coordinated hand and wrist motions during a real-time path tracing task. Approach. A two-degree-of-freedom computational musculoskeletal model was implemented for real-time EMG-driven control of a stick figure hand displayed on a computer screen. After 5–10 minutes of undirected practice, subjects were given three attempts to trace 10 straight paths, one at a time, with the fingertip of the virtual hand. Able-bodied subjects completed the task on two separate test days. Main results. Across subjects and test days, there was a significant linear relationship between log-transformed measures of accuracy and speed (Pearson’s r  =  0.25, p  <  0.0001). The amputee subject could coordinate movement between the wrist and MCP joints, but favored metacarpophalangeal joint motion more highly than able-bodied subjects in 8 of 10 trials. For able-bodied subjects, tracing accuracy was lower at the extremes of the model’s range of motion, though there was no apparent relationship between tracing accuracy and fingertip location for the amputee. Our result suggests that, unlike able-bodied subjects, the amputee’s motor control patterns were not accustomed to the multi-joint dynamics of the wrist and hand, possibly as a result of post-amputation cortical plasticity, disuse, or sensory deficits. Significance. To our knowledge, our study is one of very few that have demonstrated the real-time simultaneous control of multi-joint movements, especially wrist and finger movements, using an EMG-driven musculoskeletal model, which differs from the many data-driven algorithms that dominate the literature on EMG-driven prosthesis control. Real-time control was achieved with very little training and simple, quick (~15 s) calibration. Thus, our model is potentially a practical and effective control platform for multifunctional myoelectric prostheses that could restore more life-like hand function for individuals with upper limb amputation.}, number={3}, journal={JOURNAL OF NEURAL ENGINEERING}, author={Crouch, Dustin L. and Huang, He}, year={2017}, month={Jun} }
 @inproceedings{zhang_tran_huang_2017, title={NREL-Exo: A 4-DoFs wearable hip exoskeleton for walking and balance assistance in locomotion}, volume={2017-September}, ISBN={9781538626825}, url={http://dx.doi.org/10.1109/iros.2017.8202201}, DOI={10.1109/iros.2017.8202201}, abstractNote={In this paper, we presented a high-power, self-balancing, passively and software-controlled active compliant, and wearable hip exoskeleton to provide walking and balance assistance. The device features powered hip abduction/adduction (HAA) and hip flexion/extension (HFE) modules to provide assistance in both sagittal and frontal planes. Each module's actuation unit employs a Series Elastic Actuator (SEA) to achieve passive compliance. The hip exoskeleton can work in two basic operation modes: human-in-charge and robot-in-charge. Both modes are integrated into the low-level controller based on the admittance control, making transitions smooth and stable. A new balance controller based on the “extrapolated center of mass” (XCoM) concept is presented for real-time control hip abduction/adduction to keep the center of mass (CoM) within the support polygon. The exoskeleton controller is designed to encourage participation in walking instead of overriding users' intrinsic behavior to achieve effective assistance and training. Our preliminary experiments on a healthy subject using the hip exoskeleton demonstrated the potential effectiveness of the device and controller in assisting locomotion.}, booktitle={2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)}, publisher={IEEE}, author={Zhang, Ting and Tran, Minh and Huang, He Helen}, year={2017}, month={Sep}, pages={508–513} }
 @article{white_zhang_winslow_zahabi_zhang_huang_kaber_2017, title={Usability Comparison of Conventional Direct Control Versus Pattern Recognition Control of Transradial Prostheses}, volume={47}, ISSN={["2168-2305"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85032257100&partnerID=MN8TOARS}, DOI={10.1109/thms.2017.2759762}, abstractNote={The goal of this study was to compare the usability of two control schemes for a transradial myoelectric prosthesis, including conventional direct control (DC) and pattern recognition (PR) control, when used by able-bodied individuals. Three types of response measures were captured to assess the control schemes, including learnability, performance, and cognitive workload. Prior research has applied performance and cognitive workload metrics for evaluation of prosthetics; however, workload measures applied in these studies (e.g., heart rate, electroencephalography, and respiration rate) have many limitations. This study used eye tracking to compare cognitive load implications of the different control schemes for a two degrees-of-freedom myoelectric prosthesis. In total, 12 participants were assigned to either control condition (six persons each) or perform a clothespin relocation task. Results revealed the PR scheme to be more intuitive for users and superior to DC across all response measures. We observed a lower learning percentage (i.e., greater learning potential), lower cognitive load, and greater productivity in task performance. This preliminary study illustrates efficacy of using eye-tracking-based measures of cognitive load and standardize test paradigms for assessment of upper limb prosthetic usability and supports PR prosthetic device control as an intuitive alternative to DC.}, number={6}, journal={IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS}, author={White, Melissa Mae and Zhang, Wenjuan and Winslow, Anna T. and Zahabi, Maryam and Zhang, Fan and Huang, He and Kaber, David B.}, year={2017}, month={Dec}, pages={1146–1157} }
 @inproceedings{wen_liu_si_huang_2016, title={Adaptive control of powered transfemoral prostheses based on adaptive dynamic programming}, volume={2016-October}, ISBN={9781457702204}, url={http://dx.doi.org/10.1109/embc.2016.7591867}, DOI={10.1109/embc.2016.7591867}, abstractNote={In this study, we developed and tested a novel adaptive controller for powered transfemoral prostheses. Adaptive dynamic programming (ADP) was implemented within the prosthesis control to complement the existing finite state impedance control (FS-IC) in a prototypic active-transfemoral prosthesis (ATP). The ADP controller interacts with the human user-prosthesis system, observes the prosthesis user's dynamic states during walking, and learns to personalize user performance properties via online adaptation to meet the individual user's objectives. The new ADP controller was preliminarily tested on one able-bodied subject walking on a treadmill. The test objective was for the user to approach normative knee kinematics by tuning the FS-IC impedance parameters via ADP. The results showed the ADP was able to adjust the prosthesis controller to generate the desired normative knee kinematics within 10 minutes. In the meantime, the FS-IC impedance parameters converged at the end of the adaptive tuning procedure while maintaining the desired human-prosthesis performance. This study demonstrated the feasibility of ADP for adaptive control of a powered lower limb prosthesis. Future research efforts will address several important issues in order to validate the system on amputees. To achieve this goal, human user-centered performance objective functions will be developed, tested, and used in this adaptive controller design.}, booktitle={2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Wen, Yue and Liu, Ming and Si, Jennie and Huang, He Helen}, year={2016}, month={Aug}, pages={5071–5074} }
 @inproceedings{zhang_white_zahabi_winslow_zhang_huang_kaber_2017, title={Cognitive workload in conventional direct control vs. pattern recognition control of an upper-limb prosthesis}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85015802131&partnerID=MN8TOARS}, DOI={10.1109/SMC.2016.7844587}, abstractNote={The purpose of this study was to compare the cognitive workload of able-bodied individuals when using a myoelectric prosthetic under direct control (DC) or electromyography pattern recognition (PR) control. Different from existing clinical evaluations involving dual-task performance, pupillography measured with an eye-tracking system was used to quantitatively assess user cognitive workload in using a 2 degree-of-freedom prosthesis for a clothespin task. Test results revealed the PR control to produce superior task performance and to require lower cognitive load than demanded of participants under the DC condition. This study provided evidence of both performance and workload advantages of PR control over DC control. PR control was more intuitive to the prosthesis user and, therefore, required less cognitive effort. Furthermore, the study identified a new effective measure of cognitive workload in upper limb prosthesis use via pupillography.}, booktitle={2016 IEEE International Conference on Systems, Man, and Cybernetics, SMC 2016 - Conference Proceedings}, author={Zhang, W. J. and White, M. and Zahabi, M. and Winslow, A. T. and Zhang, F. and Huang, He and Kaber, D.}, year={2017}, pages={2335–2340} }
 @inproceedings{winslow_brantley_zhu_vidal_huang_2016, title={Corticomuscular coherence variation throughout the gait cycle during overground walking and ramp ascent: A preliminary investigation}, volume={2016-October}, ISBN={9781457702204}, url={http://dx.doi.org/10.1109/embc.2016.7591760}, DOI={10.1109/embc.2016.7591760}, abstractNote={Recent designs of neural-machine interfaces (NMIs) incorporating electroencephalography (EEG) or electromyography (EMG) have been used in lower limb assistive devices. While the results of previous studies have shown promise, a NMI which takes advantage of early movement-related EEG activity preceding movement onset, as well as the improved signal-to-noise ratio of EMG, could prove to be more accurate and responsive than current NMI designs based solely on EEG or EMG. Previous studies have demonstrated that the activity of the sensorimotor cortex is coupled to the firing rate of motor units in lower limb muscles during voluntary contraction. However, the exploration of corticomuscular coherence during locomotive tasks has been limited. In this study, coupling between the motor cortex and right tibialis anterior muscle activity was preliminarily investigated during self-paced over-ground walking and ramp ascent. EEG at the motor cortex and surface EMG from the tibialis anterior were collected from one able-bodied subject. Coherence between the two signals was computed and studied across gait cycles. The EEG activity led the EMG activity in the low gamma band in swing phase of level ground walking and in stance phase of ramp ascent. These results may inform the future design of EEG-EMG multimodal NMIs for lower limb devices that assist locomotion of people with physical disabilities.}, booktitle={2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Winslow, Anna T and Brantley, Justin and Zhu, Fangshi and Vidal, Jose L Contreras and Huang, He}, year={2016}, month={Aug}, pages={4634–4637} }
 @inproceedings{brandt_liu_huang_2016, title={Does the impedance of above-knee powered prostheses need to be adjusted for load-carrying conditions?}, volume={2016-October}, ISBN={9781457702204}, url={http://dx.doi.org/10.1109/embc.2016.7591868}, DOI={10.1109/embc.2016.7591868}, abstractNote={Powered knee prostheses provide substantial advantages for amputees compared to traditional passive devices during basic walking tasks (i.e. level-ground, stairs, ramps), but the impedance control parameters are fixed. For environments that differ from the well-controlled setting of the clinic, amputees must compensate their gait patterns because fixed control parameters ideal for walking on level ground in the clinic do not meet real-life task demands. Load carriage is one instance where fixed control parameters may lead to undesired gait patterns and potentially result in injury. To evaluate the importance of impedance control parameters for different walking tasks, we tested one above-knee amputee walking using an experimental powered prosthesis under four walking conditions. The amputee walked with and without added mass with both load-specific and non-specific impedance control parameters. The load-specific parameters significantly reduced the amputee's intact-leg compensations, asymmetry, and perceived exertion compared to the non-specific control parameters. Powered lower limb prostheses that modulate impedance control parameters for load-carrying tasks may improve the gait performance, safety, and comfort of amputees.}, booktitle={2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Brandt, Andrea and Liu, Ming and Huang, He Helen}, year={2016}, month={Aug}, pages={5075–5078} }
 @inproceedings{liu_bohlen_huang_2016, title={Effect of environmental factors on level of trip disturbance: A simulation study}, volume={2016-October}, ISBN={9781457702204}, url={http://dx.doi.org/10.1109/embc.2016.7591859}, DOI={10.1109/embc.2016.7591859}, abstractNote={Above knee amputees exhibit a higher risk of falling than able-bodied people, so the capacity to recover from trips (a major cause of unintentional falls) is critical for these amputees to prevent fall-related injuries. Although trip recovery approaches using powered prostheses have been proposed, the effectiveness of these approaches has not been evaluated with varied trip-related disturbance levels. Here, we conducted a simulation study to understand the relationship between trip-related disturbance levels and environmental factors. This knowledge could clarify the design space as well as guide design and evaluation techniques of future trip recovery approaches.}, booktitle={2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Liu, Ming and Bohlen, Peter and Huang, He Helen}, year={2016}, month={Aug}, pages={5038–5041} }
 @article{crouch_huang_2016, title={Lumped-parameter electromyogram-driven musculoskeletal hand model: A potential platform for real-time prosthesis control}, volume={49}, ISSN={["1873-2380"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85005810502&partnerID=MN8TOARS}, DOI={10.1016/j.jbiomech.2016.10.035}, abstractNote={Simple, lumped-parameter musculoskeletal models may be more adaptable and practical for clinical real-time control applications, such as prosthesis control. In this study, we determined whether a lumped-parameter, EMG-driven musculoskeletal model with four muscles could predict wrist and metacarpophalangeal (MCP) joint flexion/extension. Forearm EMG signals and joint kinematics were collected simultaneously from 5 able-bodied (AB) subjects. For one subject with unilateral transradial amputation (TRA), joint kinematics were collected from the sound arm during bilateral mirrored motion. Twenty-two model parameters were optimized such that joint kinematics predicted by EMG-driven forward dynamic simulation closely matched measured kinematics. Cross validation was employed to evaluate the model kinematic predictions using Pearson׳s correlation coefficient (r). Model predictions of joint angles were highly to very highly positively correlated with measured values at the wrist (AB mean r=0.94, TRA r=0.92) and MCP (AB mean r=0.88, TRA r=0.93) joints during single-joint wrist and MCP movements, respectively. In simultaneous multi-joint movement, the prediction accuracy for TRA at the MCP joint decreased (r=0.56), while r-values derived from AB subjects and TRA wrist motion were still above 0.75. Though parameters were optimized to match experimental sub-maximal kinematics, passive and maximum isometric joint moments predicted by the model were comparable to reported experimental measures. Our results showed the promise of a lumped-parameter musculoskeletal model for hand/wrist kinematic estimation. Therefore, the model might be useful for EMG control of powered upper limb prostheses, but more work is needed to demonstrate its online performance.}, number={16}, journal={JOURNAL OF BIOMECHANICS}, author={Crouch, Dustin L. and Huang, He}, year={2016}, month={Dec}, pages={3901–3907} }
 @inproceedings{brantley_luu_ozdemir_zhu_winslow_huang_contreras-vidal_2016, title={Noninvasive EEG correlates of overground and stair walking}, volume={2016-October}, ISBN={9781457702204}, url={http://dx.doi.org/10.1109/embc.2016.7592028}, DOI={10.1109/embc.2016.7592028}, abstractNote={Automated walking intention detection remains a challenge in lower-limb neuroprosthetic systems. Here, we assess the feasibility of extracting motor intent from scalp electroencephalography (EEG). First, we evaluated the corticomuscular coherence between central EEG electrodes (C1, Cz, C2) and muscles of the shank and thigh during walking on level ground and stairs. Second, we trained decoders to predict the linear envelope of the surface electromyogram (EMG). We observed significant EEG-led corticomuscular coupling between electrodes and sEMG (tibialis anterior) in the high delta (3-4 Hz) and low theta (4-5 Hz) frequency bands during level walking, indicating efferent signaling from the cortex to peripheral motor neurons. The coherence was increased between EEG and vastus lateralis and tibialis anterior in the delta band (<; 2 Hz) during stair ascent, indicating a task specific modulation in corticomuscular coupling. However, EMG was the leading signal for biceps femoris and gastrocnemius coherence during stair ascent, possibly representing afferent feedback loops from periphery to the motor cortex. Decoder validation showed that EEG signals contained information about the sEMG patterns during over ground walking, however, the accuracy of the predicted sEMG patterns decreased during the stair condition. Overall, these initial findings support the feasibility of integrating sEMG and EEG into a hybrid decoder for volitional control of lower limb neuroprostheses.}, booktitle={2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Brantley, Justin A. and Luu, Trieu Phat and Ozdemir, Recep and Zhu, Fangshi and Winslow, Anna T. and Huang, Helen and Contreras-Vidal, Jose L.}, year={2016}, month={Aug}, pages={5729–5732} }
 @article{zhang_bohlen_lewek_huang_2017, title={Prediction of Intrinsically Caused Tripping Events in Individuals With Stroke}, volume={25}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-85029147114&partnerID=MN8TOARS}, DOI={10.1109/tnsre.2016.2614521}, abstractNote={This study investigated the feasibility of predicting intrinsically caused trips (ICTs) in individuals with stroke. Gait kinematics collected from 12 individuals with chronic stroke, who demonstrated ICTs in treadmill walking, were analyzed. A prediction algorithm based on the outlier principle was employed. Sequential forward selection (SFS) and minimum-redundancy-maximum-relevance (mRMR) were used separately to identify the precursors for accurate ICT prediction. The results showed that it was feasible to predict ICTs around 50–260 ms before ICTs occurred in the swing phase by monitoring lower limb kinematics during the preceding stance phase. Both SFS and mRMR were effective in identifying the precursors of ICTs. For 9 out of the 12 subjects, the paretic lower limb’s shank orientation in the sagittal plane and the vertical velocity of the paretic foot’s center of gravity were important in predicting ICTs accurately; the averaged area under receiver operating characteristic curve achieved 0.95 and above. For the other three subjects, kinematics of the less affected limb or proximal joints in the paretic side were identified as the precursors to an ICT, potentially due to the variations of neuromotor deficits among stroke survivors. Although additional engineering efforts are still needed to address the challenges in making our design clinically practical, the outcome of this study may lead to further proactive engineering mechanisms for ICT avoidance and therefore reduce the risk of falls in individuals with stroke.}, number={8}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Zhang, Fan and Bohlen, Peter and Lewek, Michael D. and Huang, He}, year={2017}, month={Aug}, pages={1202–1210} }
 @inproceedings{crouch_huang_2016, title={Simple EMG-driven musculoskeletal model enables consistent control performance during path tracing tasks}, volume={2016-October}, ISBN={9781457702204}, url={http://dx.doi.org/10.1109/embc.2016.7590625}, DOI={10.1109/embc.2016.7590625}, abstractNote={Consistent, robust performance is critical for the utility and user-acceptance of neurally-controlled powered upper limb prostheses. We preliminarily evaluated the performance consistency of an electromyography (EMG)-driven controller based on a two degree-of-freedom musculoskeletal hand model, whose simplified structure is more practical for real-time prosthesis control than existing, complex models. Parameters of four virtual muscles were computed by numerical optimization from an able-bodied subject's kinematic and EMG data collected during wrist and metacarpophalangeal (MCP) flexion/extension movements. The subject attempted to trace a series of paths of different complexity (straight and curved) with the fingertip of a virtual hand displayed on a computer screen; the straight-path tracing tasks were repeated on a second test day to evaluate performance consistency over time. The subject's tracing accuracy during the tasks was consistent both between tasks of varying complexity (i.e. straight vs curved) and between test days when tracing the straight paths. Additionally, task duration, straightness, and smoothness did not significantly differ between the two straight-path test days. The consistent performance between days was achieved even with a very short (~15 seconds) calibration period to re-normalize EMG. The subject also coordinated movements of the wrist and MCP joints simultaneously during the task, much like with healthy, intact limb movement. Our promising results suggest that a musculoskeletal model-based controller may provide consistent and effective performance across a range of operating conditions, making it potentially practical for prosthesis control. Further research is needed to determine whether musculoskeletal model-based control (1) is effective for executing real-world tasks, and (2) can be extended to populations with neuromuscular impairment (e.g. amputation).}, booktitle={2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Crouch, Dustin and Huang, He}, year={2016}, month={Aug}, pages={1–4} }
 @inproceedings{zhang_liu_huang_2016, title={Tolerance of neural decoding errors for powered artificial legs: A pilot study}, volume={2016-October}, ISBN={9781457702204}, url={http://dx.doi.org/10.1109/embc.2016.7591759}, DOI={10.1109/embc.2016.7591759}, abstractNote={Neural-machine interface (NMI) decoding errors challenge the clinical value of neural control of powered artificial legs, because these errors can dangerously disturb the user's walking balance, cause stumbles or falls, and thus threaten the user's confidence and safety in prosthesis use. Although extensive research efforts have been made to minimize the NMI decoding error rate, none of the current approaches can completely eliminate the errors in NMI. This study aimed at improving the robustness of prosthesis control system against neural decoding errors by introducing a fault-tolerant control (FTC) strategy. A novel reconfiguration mechanism, combined with our previously developed NMI decoding error detector, was designed and implemented into our prototypical powered knee prosthesis. The control system with FTC was preliminarily tested on two transfemoral amputees when they walked with the powered prosthesis on different walking terrains. Various NMI errors were simulated when the FTC was enabled and disabled. The preliminary testing results indicated that the FTC strategy was capable of effectively counteracting the disruptive effects of simulated decoding errors by reducing the mechanical work change around the prosthetic knee joint elicited by the NMI error. The outcomes of this study may provide a potential engineering solution to make the neural control of powered artificial legs safer to use.}, booktitle={2016 38th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Zhang, Fan and Liu, Ming and Huang, He}, year={2016}, month={Aug}, pages={4630–4633} }
 @article{huang_crouch_liu_sawicki_wang_2016, title={A cyber expert system for auto-tuning powered prosthesis impedance control parameters}, volume={44}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84944521097&partnerID=MN8TOARS}, DOI={10.1007/s10439-015-1464-7}, abstractNote={Typically impedance control parameters (e.g., stiffness and damping) in powered lower limb prostheses are fine-tuned by human experts (HMEs), which is time and resource intensive. Automated tuning procedures would make powered prostheses more practical for clinical use. In this study, we developed a novel cyber expert system (CES) that encoded HME tuning decisions as computer rules to auto-tune control parameters for a powered knee (passive ankle) prosthesis. The tuning performance of CES was preliminarily quantified on two able-bodied subjects and two transfemoral amputees. After CES and HME tuning, we observed normative prosthetic knee kinematics and improved or slightly improved gait symmetry and step width within each subject. Compared to HME, the CES tuning procedure required less time and no human intervention. Hence, using CES for auto-tuning prosthesis control was a sound concept, promising to enhance the practical value of powered prosthetic legs. However, the tuning goals of CES might not fully capture those of the HME. This was because we observed that HME tuning reduced trunk sway, while CES sometimes led to slightly increased trunk motion. Additional research is still needed to identify more appropriate tuning objectives for powered prosthetic legs to improve amputees' walking function.}, number={5}, journal={Annals of Biomedical Engineering}, author={Huang, He and Crouch, D. L. and Liu, M. and Sawicki, G. S. and Wang, D.}, year={2016}, pages={1613–1624} }
 @article{zhang_huang_2015, title={A real-time, practical sensor fault-tolerant module for robust EMG pattern recognition}, volume={12}, ISSN={["1743-0003"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84924197446&partnerID=MN8TOARS}, DOI={10.1186/s12984-015-0011-y}, abstractNote={Unreliability of surface EMG recordings over time is a challenge for applying the EMG pattern recognition (PR)-controlled prostheses in clinical practice. Our previous study proposed a sensor fault-tolerant module (SFTM) by utilizing redundant information in multiple EMG signals. The SFTM consists of multiple sensor fault detectors and a self-recovery mechanism that can identify anomaly in EMG signals and remove the recordings of the disturbed signals from the input of the pattern classifier to recover the PR performance. While the proposed SFTM has shown great promise, the previous design is impractical. A practical SFTM has to be fast enough, lightweight, automatic, and robust under different conditions with or without disturbances. This paper presented a real-time, practical SFTM towards robust EMG PR. A novel fast LDA retraining algorithm and a fully automatic sensor fault detector based on outlier detection were developed, which allowed the SFTM to promptly detect disturbances and recover the PR performance immediately. These components of SFTM were then integrated with the EMG PR module and tested on five able-bodied subjects and a transradial amputee in real-time for classifying multiple hand and wrist motions under different conditions with different disturbance types and levels. The proposed fast LDA retraining algorithm significantly shortened the retraining time from nearly 1 s to less than 4 ms when tested on the embedded system prototype, which demonstrated the feasibility of a nearly “zero-delay” SFTM that is imperceptible to the users. The results of the real-time tests suggested that the SFTM was able to handle different types of disturbances investigated in this study and significantly improve the classification performance when one or multiple EMG signals were disturbed. In addition, the SFTM could also maintain the system’s classification performance when there was no disturbance. This paper presented a real-time, lightweight, and automatic SFTM, which paved the way for reliable and robust EMG PR for prosthesis control.}, number={1}, journal={JOURNAL OF NEUROENGINEERING AND REHABILITATION}, author={Zhang, Xiaorong and Huang, He}, year={2015}, month={Feb} }
 @inproceedings{yang_wu_liu_huang_2015, title={An Optimization-Based Approach for Prosthesis Dynamic Modeling and Parameter Identification}, volume={1}, ISBN={9780791857243}, url={http://dx.doi.org/10.1115/dscc2015-9637}, DOI={10.1115/dscc2015-9637}, abstractNote={In this paper, we propose an effective approach to model the prosthetic leg dynamics for amputees wearing active-transfemoral prosthesis (ATP) which is self-powered. To accommodate unexpected effects of thigh on knee joints, the dynamic prosthesis model has been derived using both the thigh-knee-shank and the knee-shank configurations. Correlated with the amputee’s walking data, a nonlinear optimization problem is then formulated to identify the model parameters and the gains of the PD controller which is used to control the input torque for the ATP, while reducing measurement errors of the data. Moreover, the identified models are validated by comparing the predicted dynamics with experimental measurements. The advantages of proposed method in terms of simplicity, flexibility, and accuracy are demonstrated by the high correlation coefficients and the low root-mean-square errors.}, booktitle={Volume 1: Adaptive and Intelligent Systems Control; Advances in Control Design Methods; Advances in Non-Linear and Optimal Control; Advances in Robotics; Advances in Wind Energy Systems; Aerospace Applications; Aerospace Power Optimization; Assistive Robotics; Automotive 2: Hybrid Electric Vehicles; Automotive 3: Internal Combustion Engines; Automotive Engine Control; Battery Management; Bio Engineering Applications; Biomed and Neural Systems; Connected Vehicles; Control of Robotic Systems}, publisher={American Society of Mechanical Engineers}, author={Yang, Ting and Wu, Fen and Liu, Ming and Huang, He (Helen)}, year={2015}, month={Oct} }
 @inproceedings{zhang_liu_huang_2015, title={Detection of critical errors of locomotion mode recognition for volitional control of powered transfemoral prostheses}, volume={2015-November}, ISBN={9781424492718}, url={http://dx.doi.org/10.1109/embc.2015.7318564}, DOI={10.1109/embc.2015.7318564}, abstractNote={Combination of intrinsic control and volitional control that recognizes the user's locomotion mode (LM) has been applied to powered prosthetic legs, which enables lower limb amputees to adapt to varying terrains seamlessly. However, errors in volitional control may disrupt the user's walking balance. This study aimed to determine whether the critical errors in volitional control could be detected before they disturbed the user's walking. The positive answer might lead to more robust volitional control for powered prosthetic legs. To achieve this goal, volitional control and intrinsic control were connected hierarchically to operate a powered transfemoral prosthesis. Critical errors for recognizing the user's LM were introduced artificially when transfemoral amputees walked with the powered prosthesis. Intrinsic measurements from the prosthesis were explored first in order to select effective data sources for error detection. Then a phase-dependent outlier detector was designed and offline evaluated. The results demonstrated that the designed detector can detect the tested critical errors with a high sensitivity and a low false alarm ratio before the errors disrupted the amputees' balance. Additional engineering efforts were still necessary to test the detector on more error types and design a control strategy that can make volitional control of powered lower limb prosthesis robust and safe to use.}, booktitle={2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Zhang, Fan and Liu, Ming and Huang, He}, year={2015}, month={Aug}, pages={1128–1131} }
 @article{liu_wang_huang_2016, title={Development of an Environment-Aware Locomotion Mode Recognition System for Powered Lower Limb Prostheses}, volume={24}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84963864628&partnerID=MN8TOARS}, DOI={10.1109/tnsre.2015.2420539}, abstractNote={This paper aimed to develop and evaluate an environment-aware locomotion mode recognition system for volitional control of powered artificial legs. A portable terrain recognition (TR) module, consisting of an inertia measurement unit and a laser distance meter, was built to identify the type of terrain in front of the wearer while walking. A decision tree was used to classify the terrain types and provide either coarse or refined information about the walking environment. Then, the obtained environmental information was modeled as a priori probability and was integrated with a neuromuscular-mechanical-fusion-based locomotion mode (LM) recognition system. The designed TR module and environmental-aware LM recognition system was evaluated separately on able-bodied subjects and a transfemoral amputee online. The results showed that the TR module provided high quality environmental information: TR accuracy is above 98% and terrain transitions are detected over 500 ms before the time required to switch the prosthesis control mode. This enabled smooth locomotion mode transitions for the wearers. The obtained environmental information further improved the performance of LM recognition system, regardless of whether coarse or refined information was used. In addition, the environment-aware LM recognition system produced reliable online performance when the TR output was relatively noisy, which indicated the potential of this system to operate in unconstructed environment. This paper demonstrated that environmental information should be considered for operating wearable lower limb robotic devices, such as prosthetics and orthotics.}, number={4}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Liu, Ming and Wang, Ding and Huang, He}, year={2016}, month={Apr}, pages={434–443} }
 @article{contreras-vidal_kilicarslan_huang_grossman_2015, title={Human-Centered Design of Wearable Neuroprostheses and Exoskeletons}, volume={36}, ISSN={["0738-4602"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84970039620&partnerID=MN8TOARS}, DOI={10.1609/aimag.v36i4.2613}, abstractNote={Human‐centered design of wearable robots involves the development of innovative science and technologies that minimize the mismatch between humans' and machines' capabilities, leading to their intuitive integration and confluent interaction. Here, we summarize our human‐centered approach to the design of closed‐loop brain‐machine interfaces powered prostheses and exoskeletons that allow people to act beyond their impaired or diminished physical or sensorimotor capabilities. The goal is to develop multifunctional human‐machine interfaces with integrated diagnostic, assistive, and therapeutic functions. Moreover, these complex human‐machine systems should be effective, reliable, safe, and engaging and support the patient in performing intended actions with minimal effort and errors with adequate interaction time. To illustrate our approach, we review an example of a user‐in‐the‐loop, patient‐centered, noninvasive BMI system to a powered exoskeleton for persons with paraplegia. We conclude with a summary of challenges to the translation of these complex human‐machine systems to the end user.}, number={4}, journal={AI MAGAZINE}, author={Contreras-Vidal, Jose L. and Kilicarslan, Atilla and Huang, He and Grossman, Robert G.}, year={2015}, pages={12–22} }
 @article{zhang_liu_huang_2015, title={Investigation of Timing to Switch Control Mode in Powered Knee Prostheses during Task Transitions}, volume={10}, ISSN={1932-6203}, url={http://dx.doi.org/10.1371/journal.pone.0133965}, DOI={10.1371/journal.pone.0133965}, abstractNote={Current powered prosthetic legs require switching control modes according to the task the user is performing (e.g. level-ground walking, stair climbing, walking on slopes, etc.). To allow prosthesis users safely and seamlessly transition between tasks, it is critical to determine when to switch the prosthesis control mode during task transitions. Our previous study defined critical timings for different types of task transitions in ambulation; however, it is unknown whether it is the unique timing that allows safe and seamless transitions. The goals of this study were to (1) systematically investigate the effects of mode switch timing on the prosthesis user’s performance in task transitions, and (2) identify appropriate timing to switch the prosthesis control mode so that the users can seamlessly transition between different locomotion tasks. Five able-bodied (AB) and two transfemoral (TF) amputee subjects were tested as they wore a powered knee prosthesis. The prosthesis control mode was switched manually at various times while the subjects performed different types of task transitions. The subjects’ task transition performances were evaluated by their walking balance and success in performing seamless task transitions. The results demonstrated that there existed a time window within which switching the prosthesis control mode neither interrupted the subjects’ task transitions nor disturbed their walking balance. Therefore, the results suggested the control mode switching of a lower limb prosthesis can be triggered within an appropriate time window instead of a specific timing or an individual phase. In addition, a generalized criterion to determine the appropriate mode switch timing was proposed. The outcomes of this study could provide important guidance for future designs of neurally controlled powered knee prostheses that are safe and reliable to use.}, number={7}, journal={PLOS ONE}, publisher={Public Library of Science (PLoS)}, author={Zhang, Fan and Liu, Ming and Huang, He}, editor={McCrory, Jean L.Editor}, year={2015}, month={Jul}, pages={e0133965} }
 @inproceedings{crouch_huang_2015, title={Musculoskeletal model predicts multi-joint wrist and hand movement from limited EMG control signals}, volume={2015-November}, ISBN={9781424492718}, url={http://dx.doi.org/10.1109/embc.2015.7318565}, DOI={10.1109/embc.2015.7318565}, abstractNote={Electromyography (EMG)-driven human-machine systems permit volitional control of external devices, including powered prosthetic arms. However, current control schemes are either non-intuitive to operate or lack robustness across different arm postures and dynamics, partly because these methods did not incorporate the full knowledge of biological movement production. In this study, we developed and evaluated a new musculoskeletal model to predict hand and wrist motion based on surface EMG signals. Kinematic and EMG data were collected from an able-bodied subject while performing wrist and metacarpophalangeal (MCP) joint movements with either a fixed or random speed in two static upper limb postures. A part of data collected in one posture was used to develop the model with four virtual muscles. Four parameters were optimized for each of four muscles in one posture. The model kinematic predictions were evaluated offline using the other part of the data recorded from both postures. Mean (±SD) RMS errors in predicting the joint movement were significantly lower at the MCP joint (10.1±2.5°) than at the wrist (23.5±5.2°) (p<;0.05). At both the wrist and MCP joints, the model predicted the timing and trend of joint movements reasonably well across postures and for both simple (fixed speed, single joint) and complex (random speed, simultaneous, multi-joint) movements. The results implied that our EMG-driven musculoskeletal model was promising for predicting simultaneous joint motions without significant posture and dynamics dependency. Additional engineering efforts are still needed to improve the musculoskeletal model for various human-machine interfacing applications.}, booktitle={2015 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Crouch, Dustin L. and Huang, He}, year={2015}, month={Aug}, pages={1132–1135} }
 @inproceedings{hernandez_kane_zhang_zhang_huang_2015, title={Towards ubiquitous mobile-computing-based artificial leg control}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84941308807&partnerID=MN8TOARS}, DOI={10.1109/SYSCON.2015.7116852}, abstractNote={This paper presents a rapid prototype approach for the development of a real-time capable neural-machine-interface (NMI) for control of artificial legs based on mobile processor technology (Intel Atom™ Z530 Processor.) By effectively exploiting the architectural features of a mobile embedded CPU, we implemented a decision-making algorithm, based on neuromuscular-mechanical fusion and gait phase-dependent support vector machines (SVM) classification to meet the demanding performance constraints. To demonstrate the feasibility of a real-time mobile computing based NMI, real-time experiments were performed on an able bodied subject with window increments of 50ms. The experiments showed that the mobile computing based NMI provided fast and accurate classifications of four major human locomotion tasks (level-ground walking, stair ascent, stair descent, and standing) and a 46X speedup over an equivalent MATLAB implementation. The testing yielded accuracies of 96.31% with low power consumption. An offline analysis showed the accuracy could be increased to 98.87% with minor modifications to the application.}, booktitle={9th Annual IEEE International Systems Conference, SysCon 2015 - Proceedings}, author={Hernandez, R. and Kane, J. and Zhang, F. and Zhang, X. and Huang, H.}, year={2015}, pages={821–827} }
 @article{myers_huang_zhu_2015, title={Wearable silver nanowire dry electrodes for electrophysiological sensing}, volume={5}, ISSN={2046-2069}, url={http://dx.doi.org/10.1039/c4ra15101a}, DOI={10.1039/c4ra15101a}, abstractNote={We present wearable dry electrodes made of silver nanowires for long-term electrophysiological sensing such as electrocardiography and electromyography.}, number={15}, journal={RSC Advances}, publisher={Royal Society of Chemistry (RSC)}, author={Myers, Amanda C. and Huang, He and Zhu, Yong}, year={2015}, pages={11627–11632} }
 @inproceedings{liu_zhang_huang_2014, title={An Open and configurable embedded system for EMG pattern recognition implementation for artificial arms}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84942759143&partnerID=MN8TOARS}, DOI={10.1109/embc.2014.6944524}, abstractNote={Pattern recognition (PR) based on electromyographic (EMG) signals has been developed for multifunctional artificial arms for decades. However, assessment of EMG PR control for daily prosthesis use is still limited. One of the major barriers is the lack of a portable and configurable embedded system to implement the EMG PR control. This paper aimed to design an open and configurable embedded system for EMG PR implementation so that researchers can easily modify and optimize the control algorithms upon our designed platform and test the EMG PR control outside of the lab environments. The open platform was built on an open source embedded Linux Operating System running a high-performance Gumstix board. Both the hardware and software system framework were openly designed. The system was highly flexible in terms of number of inputs/outputs and calibration interfaces used. Such flexibility enabled easy integration of our embedded system with different types of commercialized or prototypic artificial arms. Thus far, our system was portable for take-home use. Additionally, compared with previously reported embedded systems for EMG PR implementation, our system demonstrated improved processing efficiency and high system precision. Our long-term goals are (1) to develop a wearable and practical EMG PR-based control for multifunctional artificial arms, and (2) to quantify the benefits of EMG PR-based control over conventional myoelectric prosthesis control in a home setting.}, booktitle={2014 36th annual international conference of the ieee engineering in medicine and biology society (embc)}, author={Liu, J. and Zhang, F. and Huang, He}, year={2014}, pages={4095–4098} }
 @article{zhang_liu_huang_2015, title={Effects of Locomotion Mode Recognition Errors on Volitional Control of Powered Above-Knee Prostheses}, volume={23}, ISSN={["1558-0210"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84921025654&partnerID=MN8TOARS}, DOI={10.1109/tnsre.2014.2327230}, abstractNote={Recent studies have reported various methods that recognize amputees' intent regarding locomotion modes, which is potentially useful for volitional control of powered artificial legs. However, occasional errors in locomotion mode recognition are inevitable. When these intent recognition decisions are used for volitional prosthesis control, the effects of the decision errors on the operation of the prosthesis and user's task performance is unknown. Hence, the goals of this study were to 1) systematically investigate the effects of locomotion mode recognition errors on volitional control of powered prosthetic legs and the user's gait stability, and 2) identify the critical mode recognition errors that impact safe and confident use of powered artificial legs in lower limb amputees. Five able-bodied subjects and two above-knee (AK) amputees were recruited and tested when wearing a powered AK prosthesis. Four types of locomotion mode recognition errors with different duration and at different gait phases were purposely applied to the prosthesis control. The subjects' gait stabilities were subjectively and objectively quantified. The results showed that not all of the mode recognition errors in volitional prosthesis control disturb the subjects' gait stability. The effects of errors on the user's balance depended on 1) the gait phase when the errors happened and 2) the amount of mechanical work change applied on the powered knee caused by the errors. Based on the study results, “critical errors” were defined and suggested as a new index to evaluate locomotion mode recognition algorithms for artificial legs. The outcome of this study might aid the future design of volitionally-controlled powered prosthetic legs that are reliable and safe for practice.}, number={1}, journal={IEEE TRANSACTIONS ON NEURAL SYSTEMS AND REHABILITATION ENGINEERING}, author={Zhang, Fan and Liu, Ming and Huang, He}, year={2015}, month={Jan}, pages={64–72} }
 @article{zhang_liu_harper_lee_huang_2014, title={Engineering Platform and Experimental Protocol for Design and Evaluation of a Neurally-controlled Powered Transfemoral Prosthesis}, ISSN={["1940-087X"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84940242678&partnerID=MN8TOARS}, DOI={10.3791/51059}, abstractNote={To enable intuitive operation of powered artificial legs, an interface between user and prosthesis that can recognize the user's movement intent is desired. A novel neural-machine interface (NMI) based on neuromuscular-mechanical fusion developed in our previous study has demonstrated a great potential to accurately identify the intended movement of transfemoral amputees. However, this interface has not yet been integrated with a powered prosthetic leg for true neural control. This study aimed to report (1) a flexible platform to implement and optimize neural control of powered lower limb prosthesis and (2) an experimental setup and protocol to evaluate neural prosthesis control on patients with lower limb amputations. First a platform based on a PC and a visual programming environment were developed to implement the prosthesis control algorithms, including NMI training algorithm, NMI online testing algorithm, and intrinsic control algorithm. To demonstrate the function of this platform, in this study the NMI based on neuromuscular-mechanical fusion was hierarchically integrated with intrinsic control of a prototypical transfemoral prosthesis. One patient with a unilateral transfemoral amputation was recruited to evaluate our implemented neural controller when performing activities, such as standing, level-ground walking, ramp ascent, and ramp descent continuously in the laboratory. A novel experimental setup and protocol were developed in order to test the new prosthesis control safely and efficiently. The presented proof-of-concept platform and experimental setup and protocol could aid the future development and application of neurally-controlled powered artificial legs.}, number={89}, journal={JOVE-JOURNAL OF VISUALIZED EXPERIMENTS}, author={Zhang, Fan and Liu, Ming and Harper, Stephen and Lee, Michael and Huang, He}, year={2014}, month={Jul} }
 @article{hefferman_zhang_nunnery_huang_2015, title={Integration of surface electromyographic sensors with the transfemoral amputee socket: A comparison of four differing configurations}, volume={39}, ISSN={["1746-1553"]}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-84991569117&partnerID=MN8TOARS}, DOI={10.1177/0309364613516484}, abstractNote={Background and aim: In recent years, there has been an increased interest in recording high-quality electromyographic signals from within the sockets of lower-limb amputees. However, successful recording presents major challenges to both researchers and clinicians. This article details and compares four prototypical integrated socket–sensor designs used to record electromyographic signals from within the sockets of transfemoral amputees. Technique: Four prototypical socket–sensor configurations were constructed and tested on a single transfemoral amputee asked to perform sitting/standing, stair ascent/descent, and level ground walking. The number of large-amplitude motion artifacts generated using each prototype was quantified, the amount of skin irritation documented, and the comfort level of each assembly subjectively assessed by the amputee subject. Discussion: Of the four configurations tested, the combination of a suction socket with integrated wireless surface electrodes generated the lowest number of large-amplitude motion artifacts, the least visible skin irritation, and was judged to be most comfortable by the amputee subject. Clinical relevance The collection of high-quality electromyographic signals from an amputee’s residual limb while maximizing patient comfort holds substantial potential to enhance neuromuscular clinical assessment and as a method of intuitive control of powered lower-limb prostheses.}, number={2}, journal={PROSTHETICS AND ORTHOTICS INTERNATIONAL}, author={Hefferman, Gerald M. and Zhang, Fan and Nunnery, Michael J. and Huang, He}, year={2015}, month={Apr}, pages={166–173} }
 @inproceedings{myers_du_huang_zhu_2014, title={Novel wearable EMG sensors based on nanowire technology}, ISBN={9781424479290}, url={http://dx.doi.org/10.1109/embc.2014.6943928}, DOI={10.1109/embc.2014.6943928}, abstractNote={Wearable electrodes made of silver nanowires (AgNWs) have demonstrated great potential for sensing a variety of physical and physiological signals. This paper aimed to study the feasibility of AgNWs electrodes for measuring surface electromyographic (sEMG) signals. One human subject was recruited and instructed to perform wrist extension repetitively or to produce no movement in the experiment. sEMG signals were collected from the right extensor digitorum communis of the human subject by an AgNWs electrode and a commercially available Ag/AgCl wet sEMG electrode, separately. The quality of recorded sEMG in time and frequency domains was compared between the two types of electrodes. The results showed that the sEMG signals recorded by the AgNW electrode were comparable with that by the Ag/AgCl electrode. Since the dry AgNWs electrodes are flexible, wearable, and potentially robust for daily use, novel AgNW-based EMG electrodes are promising for many biomedical applications, such as myoelectric control of artificial limbs.}, booktitle={2014 36th Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Myers, Amanda and Du, Lin and Huang, He and Zhu, Yong}, year={2014}, month={Aug}, pages={1674–1677} }
 @inproceedings{zhang_huang_2014, title={Practical implementation of robust sensor interface for EMG pattern recognition for artificial arm control}, author={Zhang, F. and Huang, H.}, year={2014} }
 @inproceedings{zhang_huang_2013, title={Decoding movement intent of patient with multiple sclerosis for the powered lower extremity exoskeleton}, ISBN={9781457702167}, url={http://dx.doi.org/10.1109/embc.2013.6610660}, DOI={10.1109/embc.2013.6610660}, abstractNote={This study aims to recognize movement intent of patients with multiple sclerosis (MS) by decoding neuromuscular control signals fused with mechanical measurements as a method of powered lower extremity exoskeleton control. Surface electromyographic (EMG) signals recorded from the lower extremity muscles, ground reaction forces measured from beneath both feet, and kinematics from both thigh segments of a single MS patient were used to identify three activities (level-ground walking, sitting, and standing). Our study showed that during activity performance clear modulation of muscle activity in the lower extremities was observed for the MS patient, whose Kurtzke Expanded Disability Status Scale (EDSS) was 6. The designed intent recognition algorithm can accurately classify the subject's intended movements with 98.73% accuracy in static states and correctly predict the activity transitions about 100 to 130 ms before the actual transitions were made. These promising results indicate the potential of designed intent recognition interface for volitional control of powered lower extremity exoskeletons.}, booktitle={2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Zhang, Fan and Huang, He}, year={2013}, month={Jul}, pages={4957–4960} }
 @inproceedings{hernandez_yang_huang_zhang_zhang_2013, title={Design and implementation of a low power mobile CPU based embedded system for artificial leg control}, ISBN={9781457702167}, url={http://dx.doi.org/10.1109/embc.2013.6610862}, DOI={10.1109/embc.2013.6610862}, abstractNote={This paper presents the design and implementation of a new neural-machine-interface (NMI) for control of artificial legs. The requirements of high accuracy, real-time processing, low power consumption, and mobility of the NMI place great challenges on the computation engine of the system. By utilizing the architectural features of a mobile embedded CPU, we are able to implement our decision-making algorithm, based on neuromuscular phase-dependant support vector machines (SVM), with exceptional accuracy and processing speed. To demonstrate the superiority of our NMI, real-time experiments were performed on an able bodied subject with a 20ms window increment. The 20ms testing yielded accuracies of 99.94% while executing our algorithm efficiently with less than 11% processor loads.}, booktitle={2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Hernandez, Robert and Yang, Qing and Huang, He and Zhang, Fan and Zhang, Xiaorong}, year={2013}, month={Jul}, pages={5769–5772} }
 @inproceedings{wang_liu_zhang_huang_2013, title={Design of an expert system to automatically calibrate impedance control for powered knee prostheses}, ISBN={9781467360241 9781467360227}, url={http://dx.doi.org/10.1109/icorr.2013.6650442}, DOI={10.1109/icorr.2013.6650442}, abstractNote={Many currently available powered knee prostheses (PKP) use finite state impedance control to operate a prosthetic knee joint. The desired impedance values were usually manually calibrated with trial-and-error in order to enable near-normal walking pattern. However, such a manual approach is inaccurate, time consuming, and impractical. This paper aimed to design an expert system that can tune the control impedance for powered knee prostheses automatically and quickly. The expert system was designed based on fuzzy logic inference (FLI) to match the desired knee motion and gait timing while walking. The developed system was validated on an able-bodied subject wearing a powered prosthesis. Preliminary experimental results demonstrated that the developed expert system can converge the user's knee profile and gait timing to the desired values within 2 minutes. Additionally, after the auto-tuning procedure, the user produced more symmetrical gait. These preliminary results indicate the promise of the designed expert system for quick and accuracy impedance calibration, which can significantly improve the practical value of powered lower limb prosthesis. Continuous engineering efforts are still needed to determine the calibration objectives and validate the expert system.}, booktitle={2013 IEEE 13th International Conference on Rehabilitation Robotics (ICORR)}, publisher={IEEE}, author={Wang, Ding and Liu, Ming and Zhang, Fan and Huang, He}, year={2013}, month={Jun} }
 @article{liu_zhang_datseris_huang_2014, title={Improving Finite State Impedance Control of Active-Transfemoral Prosthesis Using Dempster-Shafer Based State Transition Rules}, volume={76}, ISSN={0921-0296 1573-0409}, url={http://dx.doi.org/10.1007/S10846-013-9979-3}, DOI={10.1007/s10846-013-9979-3}, number={3-4}, journal={Journal of Intelligent & Robotic Systems}, publisher={Springer Science and Business Media LLC}, author={Liu, Ming and Zhang, Fan and Datseris, Philip and Huang, He}, year={2014}, month={Dec}, pages={461–474} }
 @inproceedings{du_zhang_he_huang_2013, title={Improving the performance of a neural-machine interface for prosthetic legs using adaptive pattern classifiers}, ISBN={9781457702167}, url={http://dx.doi.org/10.1109/embc.2013.6609814}, DOI={10.1109/embc.2013.6609814}, abstractNote={Pattern classification has been used for design of neural-machine interface (NMI) that identifies user intent. Our previous NMI based on electromyographic (EMG) signals and intrinsic mechanical feedback has shown great promise for neural control of artificial legs. In order to make this NMI practical, however, it is desired that classification algorithms can adapt to EMG pattern variations over time, caused by various physical and physiological changes. This study aimed to develop an adaptive pattern recognition framework in the NMI to improve the robustness of NMI performance over time. Two adaptive algorithms, i.e. entropy-based adaptation and Learning From Testing Data (LIFT) adaptation, were presented and compared to the NMI with non-adaptive classifiers. Support vector machine (SVM) was selected as the basic classifier. Gradual change of EMG signals was simulated over time on EMG data collected from four transfemoral (TF) amputees. The preliminary results showed that the NMI with adaptive classifiers produced more consistent performance over time than the classifier without adaptation. The results of this preliminary study indicate the potential of using adaptive classifiers to improve the NMI reliability for neural control of powered prosthetic legs.}, booktitle={2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Du, Lin and Zhang, Fan and He, Haibo and Huang, He}, year={2013}, month={Jul}, pages={1571–1574} }
 @inproceedings{zhang_huang_yang_2013, title={Real-time implementation of a self-recovery EMG pattern recognition interface for artificial arms}, ISBN={9781457702167}, url={http://dx.doi.org/10.1109/embc.2013.6610901}, DOI={10.1109/embc.2013.6610901}, abstractNote={EMG pattern classification has been widely studied for decoding user intent for intuitive prosthesis control. However, EMG signals can be easily contaminated by noise and disturbances, which may degrade the classification performance. This study aims to design a real-time self-recovery EMG pattern classification interface to provide reliable user intent recognition for multifunctional prosthetic arm control. A novel self-recovery module consisting of multiple sensor fault detectors and a fast LDA classifier retraining strategy has been developed to immediately recover the classification performance from signal disturbances. The self-recovery EMG pattern recognition (PR) system has been implemented on an embedded system as a working prototype. Experimental evaluation has been performed on an able-bodied subject in real-time to classify three arm movements while signal disturbances were manually introduced. The results of this study may propel the clinical use of EMG PR for multifunctional prosthetic arm control.}, booktitle={2013 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC)}, publisher={IEEE}, author={Zhang, Xiaorong and Huang, He and Yang, Qing}, year={2013}, month={Jul}, pages={5926–5929} }
 @inproceedings{wang_du_huang_2013, title={Terrain recognition improves the performance of neural-machine interface for locomotion mode recognition}, ISBN={9781467352888 9781467352871 9781467352864}, url={http://dx.doi.org/10.1109/iccnc.2013.6504059}, DOI={10.1109/iccnc.2013.6504059}, abstractNote={Neural-machine interface (NMI) for artificial limbs is a typical biomedical CPS that requires seamless integration of cyber components with physical systems (i.e. prostheses and users). In this paper we aimed to adopt a bio-inspired concept to improve the performance of a NMI for artificial legs by introducing additional information about the walking environment ahead of the prosthesis user. First, a terrain recognition module based on a portable laser distance sensor and an inertial measurement unit (IMU) was designed to accurately classify the terrain type in front of the prosthesis user. The output of this module was then modeled as prior probability and integrated into a Bayesian-based NMI system. The cyber algorithms were real-time implemented and evaluated on an able-bodied subject wearing a passive prosthetic leg in the laboratory environment. The preliminary results showed that the terrain recognition module can accurately recognize the type of terrain in front of the user, approximately half to one second before the critical timing for prosthesis control mode change. NMI with or without the terrain recognition module accurately predicted all the tested task mode transitions. However, the NMI with the terrain recognition module yielded approximately 5% higher classification accuracy rate in static state and 30~105 ms earlier prediction of mode transitions than the NMI without prior knowledge of environmental information. The preliminary results demonstrated the soundness of the bio-inspired concept and established CPS framework to further enhance the accuracy and response time of NMI for artificial leg control.}, booktitle={2013 International Conference on Computing, Networking and Communications (ICNC)}, publisher={IEEE}, author={Wang, Ding and Du, Lin and Huang, He}, year={2013}, month={Jan}, pages={87–91} }
 @inproceedings{zhang_wang_yang_huang_2012, title={An automatic and user-driven training method for locomotion mode recognition for artificial leg control}, ISBN={9781457717871 9781424441198 9781457717871}, url={http://dx.doi.org/10.1109/embc.2012.6347389}, DOI={10.1109/embc.2012.6347389}, abstractNote={Our previously developed locomotion-mode-recognition (LMR) system has provided a great promise to intuitive control of powered artificial legs. However, the lack of fast, practical training methods is a barrier for clinical use of our LMR system for prosthetic legs. This paper aims to design a new, automatic, and user-driven training method for practical use of LMR system. In this method, a wearable terrain detection interface based on a portable laser distance sensor and an inertial measurement unit (IMU) is applied to detect the terrain change in front of the prosthesis user. The mechanical measurement from the prosthetic pylon is used to detect gait phase. These two streams of information are used to automatically identify the transitions among various locomotion modes, switch the prosthesis control mode, and label the training data with movement class and gait phase in real-time. No external device is required in this training system. In addition, the prosthesis user without assistance from any other experts can do the whole training procedure. The pilot experimental results on an able-bodied subject have demonstrated that our developed new method is accurate and user-friendly, and can significantly simplify the LMR training system and training procedure without sacrificing the system performance. The novel design paves the way for clinical use of our designed LMR system for powered lower limb prosthesis control.}, booktitle={2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Zhang, Xiaorong and Wang, Ding and Yang, Qing and Huang, He}, year={2012}, month={Aug}, pages={6116–6119} }
 @inproceedings{zhang_huang_yang_2012, title={Implementing an FPGA system for real-time intent recognition for prosthetic legs}, ISBN={9781450311991}, url={http://dx.doi.org/10.1145/2228360.2228394}, DOI={10.1145/2228360.2228394}, abstractNote={This paper presents the design and implementation of a cyber physical system (CPS) for neural-machine interface (NMI) that continuously senses signals from a human neuromuscular control system and recognizes the user's intended locomotion modes in real-time. The CPS contains two major parts: a microcontroller unit (MCU) for sensing and buffering input signals and an FPGA device as the computing engine for fast decoding and recognition of neural signals. The real-time experiments on a human subject demonstrated its real-time, self-contained, and high accuracy in identifying three major lower limb movement tasks (level-ground walking, stair ascent, and standing), paving the way for truly neural-controlled prosthetic legs.}, booktitle={Proceedings of the 49th Annual Design Automation Conference on - DAC '12}, publisher={ACM Press}, author={Zhang, Xiaorong and Huang, He and Yang, Qing}, year={2012}, pages={169–175} }
 @article{zhang_liu_zhang_ren_sun_yang_huang_2012, title={On Design and Implementation of Neural-Machine Interface for Artificial Legs}, volume={8}, ISSN={1551-3203 1941-0050}, url={http://dx.doi.org/10.1109/tii.2011.2166770}, DOI={10.1109/tii.2011.2166770}, abstractNote={The quality-of-life of leg amputees can be improved dramatically by using a cyber-physical system (CPS) that controls artificial legs based on neural signals representing amputees' intended movements. The key to the CPS is the neural-machine interface (NMI) that senses electromyographic (EMG) signals to make control decisions. This paper presents a design and implementation of a novel NMI using an embedded computer system to collect neural signals from a physical system-a leg amputee, provide adequate computational capability to interpret such signals, and make decisions to identify user's intent for prostheses control in real time. A new deciphering algorithm, composed of an EMG pattern classifier and a postprocessing scheme, was developed to identify the user's intended lower limb movements. To deal with environmental uncertainty, a trust management mechanism was designed to handle unexpected sensor failures and signal disturbances. Integrating the neural deciphering algorithm with the trust management mechanism resulted in a highly accurate and reliable software system for neural control of artificial legs. The software was then embedded in a newly designed hardware platform based on an embedded microcontroller and a graphic processing unit (GPU) to form a complete NMI for real-time testing. Real-time experiments on a leg amputee subject and an able-bodied subject have been carried out to test the control accuracy of the new NMI. Our extensive experiments have shown promising results on both subjects, paving the way for clinical feasibility of neural controlled artificial legs.}, number={2}, journal={IEEE Transactions on Industrial Informatics}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Zhang, Xiaorong and Liu, Yuhong and Zhang, Fan and Ren, Jin and Sun, Yan Lindsay and Yang, Qing and Huang, He}, year={2012}, month={May}, pages={418–429} }
 @inproceedings{zhang_liu_huang_2012, title={Preliminary study of the effect of user intent recognition errors on volitional control of powered lower limb prostheses}, ISBN={9781457717871 9781424441198 9781457717871}, url={http://dx.doi.org/10.1109/embc.2012.6346538}, DOI={10.1109/embc.2012.6346538}, abstractNote={Previously developed user-intent-recognition (UIR) systems have demonstrated promising accuracy for identifying the user's locomotion mode, which is potentially useful for volitional control of powered artificial legs in ambulation. The fundamental question facing us now is whether or not the prosthesis users are safe when the UIR system is directly integrated with the intrinsic controller to operate powered artificial legs. In this preliminary study, we aimed to address this question by investigating the effect of UIR errors on the walking stability of users, wearing a UIR-controlled powered transfemoral (TF) prosthesis. First, a novel control of powered prosthesis was developed, which hierarchically connected our designed UIR system with an intrinsic controller. Three types of errors were purposely added into the UIR output at different gait phase while an able-bodied subject walked on a treadmill with the powered prosthesis. Subjective opinions were obtained to evaluate the effect of applied UIR errors on the user's walking balance. The kinematics and kinetics of the prosthetic knee were quantified while the errors occurred. The preliminary results showed that not all the UIR errors applied caused a subjective feeling of balance instability. The effects of UIR errors on the prosthesis control and user's balance depended on the gait phase when the errors happened and the amount of mechanical work applied to the knee joint caused by the errors. The results of this study could aid the future design of true bionic prostheses that enable lower limb amputees to perform various activities intuitively and safely.}, booktitle={2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Zhang, Fan and Liu, Ming and Huang, He}, year={2012}, month={Aug}, pages={2768–2771} }
 @inproceedings{hernandez_zhang_zhang_huang_yang_2012, title={Promise of a low power mobile CPU based embedded system in artificial leg control}, ISBN={9781457717871 9781424441198 9781457717871}, url={http://dx.doi.org/10.1109/embc.2012.6347178}, DOI={10.1109/embc.2012.6347178}, abstractNote={This paper presents the design and implementation of a low power embedded system using mobile processor technology (Intel Atom™ Z530 Processor) specifically tailored for a neural-machine interface (NMI) for artificial limbs. This embedded system effectively performs our previously developed NMI algorithm based on neuromuscular-mechanical fusion and phase-dependent pattern classification. The analysis shows that NMI embedded system can meet real-time constraints with high accuracies for recognizing the user's locomotion mode. Our implementation utilizes the mobile processor efficiently to allow a power consumption of 2.2 watts and low CPU utilization (less than 4.3%) while executing the complex NMI algorithm. Our experiments have shown that the highly optimized C program implementation on the embedded system has superb advantages over existing PC implementations on MATLAB. The study results suggest that mobile-CPU-based embedded system is promising for implementing advanced control for powered lower limb prostheses.}, booktitle={2012 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Hernandez, R. and Zhang, Fan and Zhang, Xiaorong and Huang, He and Yang, Qing}, year={2012}, month={Aug}, pages={5250–5253} }
 @article{zhang_huang_2013, title={Source Selection for Real-Time User Intent Recognition Toward Volitional Control of Artificial Legs}, volume={17}, ISSN={2168-2194 2168-2208}, url={http://dx.doi.org/10.1109/jbhi.2012.2236563}, DOI={10.1109/jbhi.2012.2236563}, abstractNote={Various types of data sources have been used to recognize user intent for volitional control of powered artificial legs. However, there is still a debate on what exact data sources are necessary for accurately and responsively recognizing the user's intended tasks. Motivated by this widely interested question, in this study we aimed to 1) investigate the usefulness of different data sources commonly suggested for user intent recognition and 2) determine an informative set of data sources for volitional control of prosthetic legs. The studied data sources included eight surface electromyography (EMG) signals from the residual thigh muscles of transfemoral (TF) amputees, ground reaction forces/moments from a prosthetic pylon, and kinematic measurements from the residual thigh and prosthetic knee. We then ranked and included data sources based on the usefulness for user intent recognition and selected a reduced number of data sources that ensured accurate recognition of the user's intended task by using three source selection algorithms. The results showed that EMG signals and ground reaction forces/moments were more informative than prosthesis kinematics. Nine to eleven of all the initial data sources were sufficient to maintain 95% accuracy for recognizing the studied seven tasks without missing additional task transitions in real time. The selected data sources produced consistent system performance across two experimental days for four recruited TF amputee subjects, indicating the potential robustness of the selected data sources. Finally, based on the study results, we suggested a protocol for determining the informative data sources and sensor configurations for future development of volitional control of powered artificial legs.}, number={5}, journal={IEEE Journal of Biomedical and Health Informatics}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Zhang, Fan and Huang, He}, year={2013}, month={Sep}, pages={907–914} }
 @article{du_zhang_liu_huang_2012, title={Toward Design of an Environment-Aware Adaptive Locomotion-Mode-Recognition System}, volume={59}, ISSN={0018-9294 1558-2531}, url={http://dx.doi.org/10.1109/tbme.2012.2208641}, DOI={10.1109/tbme.2012.2208641}, abstractNote={In this study, we aimed to improve the performance of a locomotion-mode-recognition system based on neuromuscular-mechanical fusion by introducing additional information about the walking environment. Linear-discriminant-analysis-based classifiers were first designed to identify a lower limb prosthesis user's locomotion mode based on electromyographic signals recorded from residual leg muscles and ground reaction forces measured from the prosthetic pylon. Nine transfemoral amputees who wore a passive hydraulic knee or powered prosthetic knee participated in this study. Information about the walking terrain was simulated and modeled as prior probability based on the principle of maximum entropy and integrated into the discriminant functions of the classifier. When the correct prior knowledge of walking terrain was simulated, the classification accuracy for each locomotion mode significantly increased and no task transitions were missed. In addition, simulated incorrect prior knowledge did not significantly reduce system performance, indicating that our design is robust against noisy and imperfect prior information. Furthermore, these observations were independent of the type of prosthesis applied. The promising results in this study may assist the further development of an environment-aware adaptive system for locomotion-mode recognition for powered lower limb prostheses or orthoses.}, number={10}, journal={IEEE Transactions on Biomedical Engineering}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Du, Lin and Zhang, Fan and Liu, Ming and Huang, He}, year={2012}, month={Oct}, pages={2716–2725} }
 @inproceedings{zhang_disanto_ren_dou_yang_huang_2011, title={A Novel CPS System for Evaluating a Neural-Machine Interface for Artificial Legs}, ISBN={9781612846408}, url={http://dx.doi.org/10.1109/iccps.2011.13}, DOI={10.1109/iccps.2011.13}, abstractNote={This paper presents a novel cyber physical system (CPS) that senses signals from two physical systems -- a human neuromuscular control system and a mechanical prosthesis -- to drive a cyber virtual reality (VR) system for the purpose of evaluating a neural-machine interface (NMI) for artificial legs. Novel cyber techniques are proposed to tackle two fundamental challenges in this CPS system: inherent computation complexity for accurately identifying user's intended movements and real-time 3D rendering of a virtual avatar and environment on the cyber system. A neuromuscular-mechanical fusion algorithm is developed to decipher user intent. The decisions are then fed into the virtual reality cyber system to drive real-time motion of an avatar emulating exactly intended movements of the user. The algorithms for intent recognition and 3D VR rendering are specifically tailored to multi-core GPUs. The designed CPS system is tested on human subjects wearing prostheses. The results have shown that fusion of neuromuscular control and mechanical information improves the accuracy for user intent classification, compared to the interface based on either neuromuscular or mechanical information alone. Additionally, we find orders of magnitude speedup of GPUs over general purpose PCs, making the real-time application possible. Our prototype implementation demonstrates the feasibility of using neuromuscular-mechanical fusion to drive virtual reality in real time, which can be an effective evaluation and training tool for leg amputees to neurally control their artificial legs.}, booktitle={2011 IEEE/ACM Second International Conference on Cyber-Physical Systems}, publisher={IEEE}, author={Zhang, Fan and DiSanto, Will and Ren, Jin and Dou, Zhi and Yang, Qing and Huang, He}, year={2011}, month={Apr}, pages={67–76} }
 @article{liu_datseris_huang_2011, title={A Prototype for Smart Prosthetic Legs-Analysis and Mechanical Design}, volume={403-408}, ISSN={1662-8985}, url={http://dx.doi.org/10.4028/www.scientific.net/amr.403-408.1999}, DOI={10.4028/www.scientific.net/AMR.403-408.1999}, abstractNote={In this paper, we designed a prototype of powered above-knee prosthesis. Compared with other prototypes available in the literature, our designed prosthetic leg employs a redundant actuator concept to overcome the challenge faced by the single-motor transmission systems. The redundant actuator also enables the prosthesis to be partially functional when the prosthesis loses power. Finally, in order to provide optimal control parameters for designed above-knee prosthesis to perform a smooth level-ground walking task, an inverse dynamic model based on Kane’s method is constructed.}, journal={Advanced Materials Research}, publisher={Trans Tech Publications, Ltd.}, author={Liu, Ming and Datseris, Philip and Huang, He Helen}, year={2011}, month={Nov}, pages={1999–2006} }
 @inproceedings{liu_datseris_huang_2011, place={New Delhi, India}, title={A prototype for smart prosthetic legs: analysis and mechanical design}, volume={1}, booktitle={Proceeding of International Conference on Control, Robotics and Cybernetics}, publisher={IEEE}, author={Liu, M. and Datseris, P. and Huang, H.}, year={2011}, pages={139–143} }
 @inproceedings{zhang_huang_yang_2011, title={A special purpose embedded system for neural machine interface for artificial legs}, ISBN={9781457715891 9781424441211 9781424441228}, url={http://dx.doi.org/10.1109/iembs.2011.6091288}, DOI={10.1109/iembs.2011.6091288}, abstractNote={This paper presents a design and implementation of a neural-machine interface (NMI) for artificial legs that can decode amputee's intended movement in real time. The newly designed NMI integrates an FPGA chip for fast processing and a microcontroller unit (MCU) with multiple on-chip analog-to-digital converters (ADCs) for real-time data sampling. The resulting embedded system is able to sample in real time 12 EMG signals and 6 mechanical signals and execute a special complex phase-dependent classifier for accurate recognition of the user's intended locomotion modes. The implementation and evaluation are based on Altera's Stratix III 3S150 FPGA device coupled with Freescale's MPC5566 MCU. The experimental results for classifying three locomotion modes (level-ground walking, stairs ascent, and stairs descent) based on data collected from an able-bodied human subject have shown acceptable performance for real-time controlling of artificial legs.}, booktitle={2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Zhang, Xiaorong and Huang, He and Yang, Qing}, year={2011}, month={Aug}, pages={5207–5210} }
 @article{huang_zhang_hargrove_dou_rogers_englehart_2011, title={Continuous Locomotion-Mode Identification for Prosthetic Legs Based on Neuromuscular–Mechanical Fusion}, volume={58}, ISSN={0018-9294 1558-2531}, url={http://dx.doi.org/10.1109/tbme.2011.2161671}, DOI={10.1109/tbme.2011.2161671}, abstractNote={In this study, we developed an algorithm based on neuromuscular-mechanical fusion to continuously recognize a variety of locomotion modes performed by patients with transfemoral (TF) amputations. Electromyographic (EMG) signals recorded from gluteal and residual thigh muscles and ground reaction forces/moments measured from the prosthetic pylon were used as inputs to a phase-dependent pattern classifier for continuous locomotion-mode identification. The algorithm was evaluated using data collected from five patients with TF amputations. The results showed that neuromuscular-mechanical fusion outperformed methods that used only EMG signals or mechanical information. For continuous performance of one walking mode (i.e., static state), the interface based on neuromuscular-mechanical fusion and a support vector machine (SVM) algorithm produced 99% or higher accuracy in the stance phase and 95% accuracy in the swing phase for locomotion-mode recognition. During mode transitions, the fusion-based SVM method correctly recognized all transitions with a sufficient predication time. These promising results demonstrate the potential of the continuous locomotion-mode classifier based on neuromuscular-mechanical fusion for neural control of prosthetic legs.}, number={10}, journal={IEEE Transactions on Biomedical Engineering}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Huang, He and Zhang, Fan and Hargrove, L. J. and Dou, Zhi and Rogers, D. R. and Englehart, K. B.}, year={2011}, month={Oct}, pages={2867–2875} }
 @inproceedings{huang_dou_zhang_nunnery_2011, title={Improving the performance of a neural-machine interface for artificial legs using prior knowledge of walking environment}, ISBN={9781457715891 9781424441211 9781424441228}, url={http://dx.doi.org/10.1109/iembs.2011.6091056}, DOI={10.1109/iembs.2011.6091056}, abstractNote={A previously developed neural-machine interface (NMI) based on neuromuscular-mechanical fusion has showed promise for recognizing user locomotion modes; however, errors of NMI during mode transitions were observed, which may challenge its real application. This study aimed to investigate whether or not the prior knowledge of walking environment could further improve the NMI performance. Linear Discriminant Analysis (LDA)-based classifiers were designed to identify user intent based on electromyographic (EMG) signals from residual muscles of leg amputees and ground reaction force (GRF) measured from the prosthetic leg. The prior knowledge of the terrain in front of the user adjusted the prior possibility in the discriminant function. Therefore, the boundaries of LDA were adaptive to the prior knowledge of the walking environment. This algorithm was evaluated on a dataset collected from one patient with a transfemoral (TF) amputation. The preliminary results showed that the NMI with adaptive prior possibilities outperformed the NMI without using the prior knowledge; it produced 98.7% accuracy for identifying tested locomotion modes, accurately predicted all the task transitions with 261–390 ms prediction time, and generated stable decision during task transitions. These results indicate the potential of using prior knowledge about walking environment to further improve the NMI for prosthetic legs.}, booktitle={2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Huang, He and Dou, Zhi and Zhang, Fan and Nunnery, M. J.}, year={2011}, month={Aug}, pages={4255–4258} }
 @inproceedings{gao_zhang_huang_2011, title={Investigation of sit-to-stand and stand-to-sit in an above knee amputee}, ISBN={9781457715891 9781424441211 9781424441228}, url={http://dx.doi.org/10.1109/iembs.2011.6091712}, DOI={10.1109/iembs.2011.6091712}, abstractNote={The objective of this pilot study is twofold: 1) to extract key factors/features in sit-to-stand and stand-to-sit (STS) performed by an above knee (AK) amputee; 2) to propose a convenient way to quantify symmetry. One male unilateral transfemoral amputee participated in the pilot study. The subject was instructed to rise in a comfortable and natural manner and conduct a series of sit-to-stand, stand-to-sit. We simultaneously measured kinematics, kinetics and muscle activities. Principal component analysis (PCA) was used to reduce the dimension and identify modes and a convenient index of STS symmetry (slope of the major axis of the error ellipse) is proposed using the insole pressure sensors. Based on the preliminary results it is recommended that kinematics and kinetics in both the sagittal and frontal planes be considered for an AK amputee performing STS. The information might be useful for further research on amputee STS.}, booktitle={2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Gao, Fan and Zhang, Fan and Huang, He}, year={2011}, month={Aug}, pages={7340–7343} }
 @inproceedings{zhang_fang_liu_huang_2011, title={Preliminary design of a terrain recognition system}, ISBN={9781457715891 9781424441211 9781424441228}, url={http://dx.doi.org/10.1109/iembs.2011.6091391}, DOI={10.1109/iembs.2011.6091391}, abstractNote={This paper aims to design a wearable terrain recognition system, which might assist the control of powered artificial prosthetic legs. A laser distance sensor and inertial measurement unit (TMU) sensors were mounted on human body. These sensors were used to identify the movement state of the user, reconstruct the geometry of the terrain in front of the user while walking, and recognize the type of terrain before the user stepped on it. Different sensor configurations were investigated and compared. The designed system was evaluated on one healthy human subject when walking on an obstacle course in the laboratory environment. The results showed that the reconstructed terrain height demonstrated clearer pattern difference among studied terrains when the laser was placed on the waist than that when the laser was mounted on the shank. The designed system with the laser on the waist accurately recognized 157 out of 160 tested terrain transitions, 300ms–2870ms before the user switched the negotiated terrains. These promising results demonstrated the potential application of the designed terrain recognition system to further improve the control of powered artificial legs.}, booktitle={2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Zhang, Fan and Fang, Zheng and Liu, Ming and Huang, He}, year={2011}, month={Aug}, pages={5452–5455} }
 @inproceedings{zhang_dou_nunnery_huang_2011, title={Real-time implementation of an intent recognition system for artificial legs}, ISBN={9781457715891 9781424441211 9781424441228}, url={http://dx.doi.org/10.1109/iembs.2011.6090822}, DOI={10.1109/iembs.2011.6090822}, abstractNote={This paper presents a real-time implementation of an intent recognition system on one transfemoral (TF) amputee. Surface Electromyographic (EMG) signals recorded from residual thigh muscles and the ground reaction forces/moments collected from the prosthetic pylon were fused to identify three locomotion modes (level-ground walking, stair ascent, and stair descent) and tasks such as sitting and standing. The designed system based on neuromuscular-mechanical fusion can accurately identify the performing tasks and predict intended task transitions of the patient with a TF amputation in real-time. The overall recognition accuracy in static states (i.e. the states when subjects continuously performed the same task) was 98.36%. All task transitions were correctly recognized 80–323 ms before the defined critical timing for safe switch of prosthesis control mode. These promising results indicate the potential of designed intent recognition system for neural control of computerized, powered prosthetic legs.}, booktitle={2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Zhang, Fan and Dou, Zhi and Nunnery, M. and Huang, He}, year={2011}, month={Aug}, pages={2997–3000} }
 @inproceedings{zhang_huang_2011, title={Real-time recognition of user intent for neural control of aritifcial legs}, author={Zhang, F. and Huang, H.}, year={2011} }
 @article{zhang_d'andrea_nunnery_kay_huang_2011, title={Towards Design of a Stumble Detection System for Artificial Legs}, volume={19}, ISSN={1534-4320 1558-0210}, url={http://dx.doi.org/10.1109/tnsre.2011.2161888}, DOI={10.1109/tnsre.2011.2161888}, abstractNote={Recent advances in design of powered artificial legs have led to increased potential to allow lower limb amputees to actively recover from stumbles. To achieve this goal, promptly and accurately identifying stumbles is essential. This study aimed to 1) select potential stumble detection data sources that react reliably and quickly to stumbles and can be measured from a prosthesis, and 2) investigate two different approaches based on selected data sources to detect stumbles and classify stumble types in patients with transfemoral (TF) amputations during ambulation. In the experiments, the normal gait of TF amputees was perturbed by a controllable treadmill or when they walked on an obstacle course. The results showed that the acceleration of prosthetic foot can accurately detect the tested stumbling events 140-240 ms before the critical timing of falling and precisely classify the stumble type. However, the detector based on foot acceleration produced high false alarm rates, which challenged its real application. Combining electromyographic (EMG) signals recorded from the residual limb with the foot acceleration significantly reduced the false alarm rate but sacrificed the detection response time. The results of this study may lead to design of a stumble detection system for instrumented, powered artificial legs; however, continued engineering efforts are required to improve the detection performance and resolve the challenges that remain for implementing the stumble detector on prosthetic legs.}, number={5}, journal={IEEE Transactions on Neural Systems and Rehabilitation Engineering}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Zhang, Fan and D'Andrea, S. E. and Nunnery, M. J. and Kay, S. M. and Huang, He}, year={2011}, month={Oct}, pages={567–577} }
 @inproceedings{liu_zhang_sun_huang_2011, title={Trust sensor interface for improving reliability of EMG-based user intent recognition}, ISBN={9781457715891 9781424441211 9781424441228}, url={http://dx.doi.org/10.1109/iembs.2011.6091853}, DOI={10.1109/iembs.2011.6091853}, abstractNote={To achieve natural and smooth control of prostheses, Electromyographic (EMG) signals have been investigated for decoding user intent. However, EMG signals can be easily contaminated by diverse disturbances, leading to errors in user intent recognition and threatening the safety of prostheses users. To address this problem, we propose a trust sensor interface (TSI) that contains 2 modules: (1) abnormality detector that detects diverse disturbances with high accuracy and low latency and (2) trust evaluation that dynamically evaluates the reliability of EMG sensors. Based on the output of the TSI, the user intention recognition (UIR) algorithm is able to dynamically adjust their operations or decisions. Our experiments on an able-bodied subject have demonstrated that the proposed TSI can effectively detect two types of disturbances (i.e. motion artifacts and baseline shifts) and improve the reliability of the UIR.}, booktitle={2011 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Liu, Yuhong and Zhang, Fan and Sun, Yan and Huang, He}, year={2011}, month={Aug}, pages={7516–7520} }
 @inbook{wolf_huang_2010, place={Philadelphia}, edition={5th}, title={Chapter 70: Evolution of Biofeedback in Physical Medicine and Rehabilitation}, booktitle={DeLisa’s Physical Medicine and Rehabilitation: Principles and Practice}, publisher={Lippincott Williams & Wilkins}, author={Wolf, S.L. and Huang, H.}, editor={Frontera, WEditor}, year={2010} }
 @inproceedings{zhang_huang_yang_2010, title={Design and implementation of a special purpose embedded system for neural machine interface}, ISBN={9781424489367}, url={http://dx.doi.org/10.1109/iccd.2010.5647801}, DOI={10.1109/iccd.2010.5647801}, abstractNote={Our previous study has shown the potential of using a computer system to accurately decode electromyographic (EMG) signals for neural controlled artificial legs. Because of computation complexity of the training algorithm coupled with real time requirement of controlling artificial legs, traditional embedded systems generally cannot be directly applied to the system. This paper presents a new design of an FPGA-based neural-machine interface for artificial legs. Both the training algorithm and the real time controlling algorithm are implemented on an FPGA. A soft processor built on the FPGA is used to manage hardware components and direct data flows. The implementation and evaluation of this design are based on Altera Stratix II GX EP2SGX90 FPGA device on a PCI Express development board. Our performance evaluations indicate that a speedup of around 280X can be achieved over our previous software implementation with no sacrifice of computation accuracy. The results demonstrate the feasibility of a self-contained, low power, and high performance real-time neural-machine interface for artificial legs.}, booktitle={2010 IEEE International Conference on Computer Design}, publisher={IEEE}, author={Zhang, Xiaorong and Huang, He and Yang, Qing}, year={2010}, month={Oct}, pages={166–172} }
 @inproceedings{lin_zhang_huang_yang_2010, title={Design and implementation of an embedded system for neural-controlled artificial legs}, ISBN={9781424449972}, url={http://dx.doi.org/10.1109/whcm.2010.5441272}, DOI={10.1109/whcm.2010.5441272}, abstractNote={This paper presents a design and partial implementation of an embedded system as a part of neural-machine interface (NMI) for neural-controlled artificial legs. We have designed a circuit consisting of 30 analog inputs for sampling signals from 16 EMG (Electromyography) electrodes, a 6 degrees of freedom (DOFs) load cell, 5 force sensitive resistors (FSR), and 3 goniometers. The amplified signals are filtered and converted to digital information, which is stored in a RAM. A special pattern recognition algorithm is then executed on the embedded CPU in association with the flash memory that stores the prior training data to make real time decisions. A preliminary prototype with one analog channel has been built and MPC5566 microcontroller has been used to implement the pattern recognition algorithm to measure the execution time. Measurement results show the feasibility of real time processing of neural controlled artificial legs.}, booktitle={2010 IEEE Workshop on Health Care Management (WHCM)}, publisher={IEEE}, author={Lin, An and Zhang, Xiaorong and Huang, He and Yang, Qing}, year={2010}, month={Feb} }
 @article{huang_zhang_sun_he_2010, title={Design of a robust EMG sensing interface for pattern classification}, volume={7}, ISSN={1741-2560 1741-2552}, url={http://dx.doi.org/10.1088/1741-2560/7/5/056005}, DOI={10.1088/1741-2560/7/5/056005}, abstractNote={Electromyographic (EMG) pattern classification has been widely investigated for neural control of external devices in order to assist with movements of patients with motor deficits. Classification performance deteriorates due to inevitable disturbances to the sensor interface, which significantly challenges the clinical value of this technique. This study aimed to design a sensor fault detection (SFD) module in the sensor interface to provide reliable EMG pattern classification. This module monitored the recorded signals from individual EMG electrodes and performed a self-recovery strategy to recover the classification performance when one or more sensors were disturbed. To evaluate this design, we applied synthetic disturbances to EMG signals collected from leg muscles of able-bodied subjects and a subject with a transfemoral amputation and compared the accuracies for classifying transitions between different locomotion modes with and without the SFD module. The results showed that the SFD module maintained classification performance when one signal was distorted and recovered about 20% of classification accuracy when four signals were distorted simultaneously. The method was simple to implement. Additionally, these outcomes were observed for all subjects, including the leg amputee, which implies the promise of the designed sensor interface for providing a reliable neural–machine interface for artificial legs.}, number={5}, journal={Journal of Neural Engineering}, publisher={IOP Publishing}, author={Huang, He and Zhang, Fan and Sun, Yan L and He, Haibo}, year={2010}, month={Sep}, pages={056005} }
 @inproceedings{huang_sun_yang_zhang_zhang_liu_ren_sierra_2010, title={Integrating neuromuscular and cyber systems for neural control of artificial legs}, ISBN={9781450300667}, url={http://dx.doi.org/10.1145/1795194.1795213}, DOI={10.1145/1795194.1795213}, abstractNote={This paper presents a design and implementation of a cyber-physical system (CPS) for neurally controlled artificial legs. The key to the new CPS system is the neural-machine interface (NMI) that uses an embedded computer to collect and interpret electromyographic (EMG) signals from a physical system that is a leg amputee. A new deciphering algorithm, composed of an EMG pattern classifier and finite state machine (FSM), was developed to identify the user's intended lower limb movements. To deal with environmental uncertainty, a trust management mechanism was designed to handle unexpected sensor failures and signal disturbances. Integrating the neural deciphering algorithm with the trust management mechanism resulted in a highly accurate and reliable software system for neural control of artificial legs. The software was then embedded in a newly designed hardware platform based on an embedded microcontroller and a graphic processing unit (GPU) to form a complete NMI for real time testing. Our preliminary experiment on a human subject demonstrated the feasibility of our designed real-time neural-machine interface for artificial legs.}, booktitle={Proceedings of the 1st ACM/IEEE International Conference on Cyber-Physical Systems - ICCPS '10}, publisher={ACM Press}, author={Huang, He and Sun, Yan (Lindsay) and Yang, Qing and Zhang, Fan and Zhang, Xiaorong and Liu, Yuhong and Ren, Jin and Sierra, Fabian}, year={2010}, pages={129–138} }
 @article{cao_he_huang_2011, title={LIFT: A new framework of learning from testing data for face recognition}, volume={74}, ISSN={0925-2312}, url={http://dx.doi.org/10.1016/j.neucom.2010.10.015}, DOI={10.1016/j.neucom.2010.10.015}, abstractNote={In this paper, a novel learning methodology for face recognition, LearnIng From Testing data (LIFT) framework, is proposed. Considering many face recognition problems featured by the inadequate training examples and availability of the vast testing examples, we aim to explore the useful information from the testing data to facilitate learning. The one-against-all technique is integrated into the learning system to recover the labels of the testing data, and then expand the training population by such recovered data. In this paper, neural networks and support vector machines are used as the base learning models. Furthermore, we integrate two other transductive methods, consistency method and LRGA method into the LIFT framework. Experimental results and various hypothesis testing over five popular face benchmarks illustrate the effectiveness of the proposed framework.}, number={6}, journal={Neurocomputing}, publisher={Elsevier BV}, author={Cao, Yuan and He, Haibo and Huang, He (Helen)}, year={2011}, month={Feb}, pages={916–929} }
 @article{tkach_huang_kuiken_2010, title={Study of stability of time-domain features for electromyographic pattern recognition}, volume={7}, ISSN={1743-0003}, url={http://dx.doi.org/10.1186/1743-0003-7-21}, DOI={10.1186/1743-0003-7-21}, abstractNote={Abstract
          
            Background
            Significant progress has been made towards the clinical application of human-machine interfaces (HMIs) based on electromyographic (EMG) pattern recognition for various rehabilitation purposes. Making this technology practical and available to patients with motor deficits requires overcoming real-world challenges, such as physical and physiological changes, that result in variations in EMG signals and systems that are unreliable for long-term use. In this study, we aimed to address these challenges by (1) investigating the stability of time-domain EMG features during changes in the EMG signals and (2) identifying the feature sets that would provide the most robust EMG pattern recognition.
          
          
            Methods
            Variations in EMG signals were introduced during physical experiments. We identified three disturbances that commonly affect EMG signals: EMG electrode location shift, variation in muscle contraction effort, and muscle fatigue. The impact of these disturbances on individual features and combined feature sets was quantified by changes in classification performance. The robustness of feature sets was evaluated by a stability index developed in this study.
          
          
            Results
            Muscle fatigue had the smallest effect on the studied EMG features, while electrode location shift and varying effort level significantly reduced the classification accuracy for most of the features. Under these disturbances, the most stable EMG feature set with combination of four features produced at least 16.0% higher classification accuracy than the least stable set. EMG autoregression coefficients and cepstrum coefficients showed the most robust classification performance of all studied time-domain features.
          
          
            Conclusions
            Selecting appropriate EMG feature combinations can overcome the impact of the studied disturbances on EMG pattern classification to a certain extent; however, this simple solution is still inadequate. Stabilizing electrode contact locations and developing effective classifier training strategies are suggested to further improve the robustness of HMIs based on EMG pattern recognition.
          }, number={1}, journal={Journal of NeuroEngineering and Rehabilitation}, publisher={Springer Science and Business Media LLC}, author={Tkach, Dennis and Huang, He and Kuiken, Todd A}, year={2010}, pages={21} }
 @inproceedings{xiao_huang_sun_yang_2009, title={Promise of embedded system with GPU in artificial leg control: Enabling time-frequency feature extraction from electromyography}, url={http://dx.doi.org/10.1109/iembs.2009.5333633}, DOI={10.1109/iembs.2009.5333633}, abstractNote={Applying electromyographic (EMG) signal pattern recognition to artificial leg control is challenging because leg EMGs are non-stationary. Time-frequency features are suitable for representing non-stationary signals; however, the computational complexity to extract time-frequency features is too high and current embedded systems used for artificial limb control are inadequate for real-time computing. The aim of this study was to quantify the computational speed of a novel embedded system, the Graphic Processor Unit (GPU), on EMG time-frequency feature extraction. The computational time derived from a GPU was compared to that derived from a general purpose CPU. The results indicated that the GPU significantly increased the computational speed. When the size of EMG analysis window was set to 100 ms, the GPU extracted EMG time-frequency features over 50 times faster than the CPU setting. Therefore, high performance GPU shows a great promise for EMG-controlled artificial legs and other medical applications that need high-speed and real-time computation.}, booktitle={2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Xiao, Weijun and Huang, He and Sun, Yan and Yang, Qing}, year={2009}, month={Sep}, pages={6926–6929} }
 @article{huang_zhou_li_kuiken_2009, title={Spatial Filtering Improves EMG Classification Accuracy Following Targeted Muscle Reinnervation}, volume={37}, ISSN={0090-6964 1573-9686}, url={http://dx.doi.org/10.1007/s10439-009-9737-7}, DOI={10.1007/s10439-009-9737-7}, abstractNote={The combination of targeted muscle reinnervation (TMR) and pattern classification of electromyography (EMG) has shown great promise for multifunctional myoelectric prosthesis control. In this study, we hypothesized that surface EMG recordings with high spatial resolution over reinnervated muscles could capture focal muscle activity and improve the classification accuracy of identifying intended movements. To test this hypothesis, TMR subjects with transhumeral or shoulder disarticulation amputations were recruited. Spatial filters such as single differential filters, double differential filters, and various two-dimensional, high-order spatial filters were used, and the classification accuracies for fifteen different movements were calculated. Compared with monopolar recordings, spatially localized EMG signals produced increased accuracy in identifying the TMR patients' movement intents, especially for hand movements. When the number of EMG signals was constrained to 12, the double differential filters gave 5-15% higher classification accuracies than the filters with lower spatial resolution, but resulted in comparable accuracies to the filters with higher spatial resolution. These results suggest that double differential EMG recordings may further improve the TMR-based neural interface for robust, multifunctional control of artificial arms.}, number={9}, journal={Annals of Biomedical Engineering}, publisher={Springer Science and Business Media LLC}, author={Huang, He and Zhou, Ping and Li, Guanglin and Kuiken, Todd}, year={2009}, month={Jun}, pages={1849–1857} }
 @inproceedings{hargrove_huang_schultz_lock_lipschutz_kuiken_2009, title={Toward the development of a neural interface for lower limb prosthesis control}, url={http://dx.doi.org/10.1109/iembs.2009.5334303}, DOI={10.1109/iembs.2009.5334303}, abstractNote={Lower limb amputees form a large portion of the amputee population; however, current lower limb prostheses do not meet the needs of patients with high-level amputations who need to perform multi-joint coordinated movements. A critical missing element is an intuitive neural interface from which user intent can be determined. Surface EMG has been used as control source for upper limb prostheses for many years; for lower limb activities, however, the EMG is non-stationary and a new control strategy is required. This paper describes the work completed to date in developing a novel lower limb neural interface.}, booktitle={2009 Annual International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Hargrove, L.J. and Huang, H. and Schultz, A.E. and Lock, B.A. and Lipschutz, R. and Kuiken, T.A.}, year={2009}, month={Sep}, pages={2111–2114} }
 @article{huang_kuiken_lipschutz_2009, title={A Strategy for Identifying Locomotion Modes Using Surface Electromyography}, volume={56}, ISSN={0018-9294 1558-2531}, url={http://dx.doi.org/10.1109/tbme.2008.2003293}, DOI={10.1109/tbme.2008.2003293}, abstractNote={This study investigated the use of surface electromyography (EMG) combined with pattern recognition (PR) to identify user locomotion modes. Due to the nonstationary characteristics of leg EMG signals during locomotion, a new phase-dependent EMG PR strategy was proposed for classifying the user's locomotion modes. The variables of the system were studied for accurate classification and timely system response. The developed PR system was tested on EMG data collected from eight able-bodied subjects and two subjects with long transfemoral (TF) amputations while they were walking on different terrains or paths. The results showed reliable classification for the seven tested modes. For eight able-bodied subjects, the average classification errors in the four defined phases using ten electrodes located over the muscles above the knee (simulating EMG from the residual limb of a TF amputee) were 12.4% plusmn 5.0%, 6.0% plusmn 4.7%, 7.5% plusmn 5.1%, and 5.2% plusmn 3.7%, respectively. Comparable results were also observed in our pilot study on the subjects with TF amputations. The outcome of this investigation could promote the future design of neural-controlled artificial legs.}, number={1}, journal={IEEE Transactions on Biomedical Engineering}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Huang, He and Kuiken, T.A. and Lipschutz, R.D.}, year={2009}, month={Jan}, pages={65–73} }
 @article{huang_zhou_li_kuiken_2008, title={An Analysis of EMG Electrode Configuration for Targeted Muscle Reinnervation Based Neural Machine Interface}, volume={16}, ISSN={1534-4320 1558-0210}, url={http://dx.doi.org/10.1109/tnsre.2007.910282}, DOI={10.1109/tnsre.2007.910282}, abstractNote={Targeted muscle reinnervation (TMR) is a novel neural machine interface for improved myoelectric prosthesis control. Previous high-density (HD) surface electromyography (EMG) studies have indicated that tremendous neural control information can be extracted from the reinnervated muscles by EMG pattern recognition (PR). However, using a large number of EMG electrodes hinders clinical application of the TMR technique. This study investigated a reduced number of electrodes and the placement required to extract sufficient neural control information for accurate identification of user movement intents. An electrode selection algorithm was applied to the HD EMG recordings from each of four TMR amputee subjects. The results show that when using only 12 selected bipolar electrodes the average accuracy over subjects for classifying 16 movement intents was 93.0 (plusmn3.3)%, just 1.2% lower than when using the entire HD electrode complement. The locations of selected electrodes were consistent with the anatomical reinnervation sites. Additionally, a practical protocol for clinical electrode placement was developed, which does not rely on complex HD EMG experiment and analysis while maintaining a classification accuracy of 88.7plusmn4.5%. These outcomes provide important guidelines for practical electrode placement that can promote future clinical application of TMR and EMG PR in the control of multifunctional prostheses.}, number={1}, journal={IEEE Transactions on Neural Systems and Rehabilitation Engineering}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Huang, He and Zhou, Ping and Li, Guanglin and Kuiken, T.A.}, year={2008}, month={Feb}, pages={37–45} }
 @article{miller_lipschutz_stubblefield_lock_huang_williams_weir_kuiken_2008, title={Control of a Six Degree of Freedom Prosthetic Arm After Targeted Muscle Reinnervation Surgery}, volume={89}, ISSN={0003-9993}, url={http://dx.doi.org/10.1016/j.apmr.2008.05.016}, DOI={10.1016/j.apmr.2008.05.016}, abstractNote={To fit and evaluate the control of a complex prosthesis for a shoulder disarticulation-level amputee with targeted muscle reinnervation.One participant who had targeted muscle reinnervation surgery was fitted with an advanced prosthesis and his use of this device was compared with the device that he used in the home setting.The experiments were completed within a laboratory setting.The first recipient of targeted muscle reinnervation: a bilateral shoulder disarticulation-level amputee.Two years after surgery, the subject was fitted with a 6 degree of freedom (DOF) prosthesis (shoulder flexion, humeral rotation, elbow flexion, wrist rotation, wrist flexion, and hand control). Control of this device was compared with that of his commercially available 3-DOF system (elbow, wrist rotation, and powered hook terminal device).In order to assess performance, movement analysis and timed movement tasks were executed.The subject was able to independently operate all 6 arm functions with good control. He could simultaneously operate 2 DOF of several different joint combinations with relative ease. He operated up to 4 DOF simultaneously, but with poor control. Work space was markedly increased and some timed tasks were faster with the 6-DOF system.This proof-of-concept study shows that advances in control of shoulder disarticulation-level prostheses can improve the quality of movement. Additional control sources may spur the development of more advanced and complex componentry for these amputees.}, number={11}, journal={Archives of Physical Medicine and Rehabilitation}, publisher={Elsevier BV}, author={Miller, Laura A. and Lipschutz, Robert D. and Stubblefield, Kathy A. and Lock, Blair A. and Huang, He and Williams, T. Walley, III and Weir, Richard F. and Kuiken, Todd A.}, year={2008}, month={Nov}, pages={2057–2065} }
 @article{zhou_lowery_englehart_huang_li_hargrove_dewald_kuiken_2007, title={Decoding a New Neural–Machine Interface for Control of Artificial Limbs}, volume={98}, ISSN={0022-3077 1522-1598}, url={http://dx.doi.org/10.1152/jn.00178.2007}, DOI={10.1152/jn.00178.2007}, abstractNote={An analysis of the motor control information content made available with a neural–machine interface (NMI) in four subjects is presented in this study. We have developed a novel NMI–called targeted muscle reinnervation (TMR)—to improve the function of artificial arms for amputees. TMR involves transferring the residual amputated nerves to nonfunctional muscles in amputees. The reinnervated muscles act as biological amplifiers of motor commands in the amputated nerves and the surface electromyogram (EMG) can be used to enhance control of a robotic arm. Although initial clinical success with TMR has been promising, the number of degrees of freedom of the robotic arm that can be controlled has been limited by the number of reinnervated muscle sites. In this study we assess how much control information can be extracted from reinnervated muscles using high-density surface EMG electrode arrays to record surface EMG signals over the reinnervated muscles. We then applied pattern classification techniques to the surface EMG signals. High accuracy was achieved in the classification of 16 intended arm, hand, and finger/thumb movements. Preliminary analyses of the required number of EMG channels and computational demands demonstrate clinical feasibility of these methods. This study indicates that TMR combined with pattern-recognition techniques has the potential to further improve the function of prosthetic limbs. In addition, the results demonstrate that the central motor control system is capable of eliciting complex efferent commands for a missing limb, in the absence of peripheral feedback and without retraining of the pathways involved.}, number={5}, journal={Journal of Neurophysiology}, publisher={American Physiological Society}, author={Zhou, Ping and Lowery, Madeleine M. and Englehart, Kevin B. and Huang, He and Li, Guanglin and Hargrove, Levi and Dewald, Julius P. A. and Kuiken, Todd A.}, year={2007}, month={Nov}, pages={2974–2982} }
 @article{sugar_he_koeneman_koeneman_herman_huang_schultz_herring_wanberg_balasubramanian_et al._2007, title={Design and Control of RUPERT: A Device for Robotic Upper Extremity Repetitive Therapy}, volume={15}, ISSN={1534-4320 1558-0210}, url={http://dx.doi.org/10.1109/tnsre.2007.903903}, DOI={10.1109/tnsre.2007.903903}, abstractNote={The structural design, control system, and integrated biofeedback for a wearable exoskeletal robot for upper extremity stroke rehabilitation are presented. Assisted with clinical evaluation, designers, engineers, and scientists have built a device for robotic assisted upper extremity repetitive therapy (RUPERT). Intense, repetitive physical rehabilitation has been shown to be beneficial overcoming upper extremity deficits, but the therapy is labor intensive and expensive and difficult to evaluate quantitatively and objectively. The RUPERT is developed to provide a low cost, safe and easy-to-use, robotic-device to assist the patient and therapist to achieve more systematic therapy at home or in the clinic. The RUPERT has four actuated degrees-of-freedom driven by compliant and safe pneumatic muscles (PMs) on the shoulder, elbow, and wrist. They are programmed to actuate the device to extend the arm and move the arm in 3-D space. It is very important to note that gravity is not compensated and the daily tasks are practiced in a natural setting. Because the device is wearable and lightweight to increase portability, it can be worn standing or sitting providing therapy tasks that better mimic activities of daily living. The sensors feed back position and force information for quantitative evaluation of task performance. The device can also provide real-time, objective assessment of functional improvement. We have tested the device on stroke survivors performing two critical activities of daily living (ADL): reaching out and self feeding. The future improvement of the device involves increased degrees-of-freedom and interactive control to adapt to a user's physical conditions.}, number={3}, journal={IEEE Transactions on Neural Systems and Rehabilitation Engineering}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Sugar, T.G. and He, Jiping and Koeneman, E.J. and Koeneman, J.B. and Herman, R. and Huang, H. and Schultz, R.S. and Herring, D.E. and Wanberg, J. and Balasubramanian, S. and et al.}, year={2007}, month={Sep}, pages={336–346} }
 @article{huang_he_herman_carhart_2006, title={Modulation Effects of Epidural Spinal Cord Stimulation on Muscle Activities During Walking}, volume={14}, ISSN={1534-4320}, url={http://dx.doi.org/10.1109/tnsre.2005.862694}, DOI={10.1109/tnsre.2005.862694}, abstractNote={Epidural spinal cord stimulation (ESCS) combined with partial weight bearing therapy (PWBT) has been reported to facilitate recovery of functional walking for individuals after chronic incomplete spinal cord injury (ISCI). Muscle activities were analyzed in this report to examine the modulation effect of ESCS on muscle recruitment during gait training. Two ISCI individuals participated in the study and both are classified as ASIA C with low motor scores in the lower limbs. Stimulating electrodes were placed at the epidural space over T10-L2 spinal segments, along the midline in participant 1 (S1), and off-midline in participant 2 (S2). Surface electromyograms (EMGs) from leg muscles under both ESCS ON and OFF conditions recorded during treadmill gait were analyzed in time-frequency domains. ESCS application produced acute modulations in muscle activities in both participants, but the observed pattern, magnitude, and spectral content of the EMGs differed. In S1, ESCS induced a significant shift in the temporal pattern of muscle activity toward normal comparing with that when ESCS was OFF, though without eliciting noticeable change in frequency distribution between ESCS ON and OFF conditions. When ESCS was applied in S2, a modulation of EMG magnitude was observed and, consequently, improved joint kinematics during walking. In this case, a stimulation entrainment appeared in time-frequency analysis. The results suggest that ESCS activates neural structures in the dorsal aspect of the spinal cord and facilitates gait-related muscle recruitment. The exact effects of ESCS depend on the electrode placement and possibly injury history and residual functions, but in general ESCS produces a positive effect on improved walking speed, endurance, and reduced sense of effort in both ISCI subjects.}, number={1}, journal={IEEE Transactions on Neural Systems and Rehabilitation Engineering}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Huang, H. and He, J. and Herman, R. and Carhart, M.R.}, year={2006}, month={Mar}, pages={14–23} }
 @inproceedings{huang_chen_xu_sundaram_olson_ingalls_rikakis_he_2006, title={Novel Design of Interactive Multimodal Biofeedback System for Neurorehabilitation}, ISBN={1424400325}, url={http://dx.doi.org/10.1109/iembs.2006.260409}, DOI={10.1109/iembs.2006.260409}, abstractNote={A previous design of a biofeedback system for Neurorehabilitation in an interactive multimodal environment has demonstrated the potential of engaging stroke patients in task-oriented neuromotor rehabilitation. This report explores the new concept and alternative designs of multimedia based biofeedback systems. In this system, the new interactive multimodal environment was constructed with abstract presentation of movement parameters. Scenery images or pictures and their clarity and orientation are used to reflect the arm movement and relative position to the target instead of the animated arm. The multiple biofeedback parameters were classified into different hierarchical levels w.r.t. importance of each movement parameter to performance. A new quantified measurement for these parameters were developed to assess the patient's performance both real-time and offline. These parameters were represented by combined visual and auditory presentations with various distinct music instruments. Overall, the objective of newly designed system is to explore what information and how to feedback information in interactive virtual environment could enhance the sensorimotor integration that may facilitate the efficient design and application of virtual environment based therapeutic intervention}, booktitle={2006 International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Huang, He and Chen, Y and Xu, W and Sundaram, H. and Olson, L. and Ingalls, T and Rikakis, T and He, Jiping}, year={2006}, month={Aug}, pages={4925–4928} }
 @inproceedings{balasubramanian_huang_he_2006, title={Quantification of Dynamic Property of Pneumatic Muscle Actuator for Design of Therapeutic Robot Control}, ISBN={1424400325}, url={http://dx.doi.org/10.1109/iembs.2006.259536}, DOI={10.1109/iembs.2006.259536}, abstractNote={Robot-assisted therapy has shown potential in neuromotor rehabilitation. A therapeutic robot driven by pneumatic muscle actuators has been developed in our research group. However, the design of fine and real-time feedback robot control is a challenge. One of the difficulties is the lack of a general dynamic model of the pneumatic muscle actuator. In this study, a phenomenological model has been developed to quantify the dynamic behavior of pneumatic muscle actuator by fitting the experimental length response of the pneumatic muscle, to a step pressure input. In addition, comparison of the dynamic responses of two pneumatic muscles of different dimensions has also been studied. Several control strategies for the pneumatic muscle actuator are discussed based on the results from this study}, booktitle={2006 International Conference of the IEEE Engineering in Medicine and Biology Society}, publisher={IEEE}, author={Balasubramanian, Sivakumar and Huang, He and He, Jiping}, year={2006}, month={Aug}, pages={2734–2737} }
 @article{huang_wolf_he_2006, title={Recent developments in biofeedback for neuromotor rehabilitation}, volume={3}, ISSN={1743-0003}, url={http://dx.doi.org/10.1186/1743-0003-3-11}, DOI={10.1186/1743-0003-3-11}, abstractNote={Abstract
          The original use of biofeedback to train single muscle activity in static positions or movement unrelated to function did not correlate well to motor function improvements in patients with central nervous system injuries. The concept of task-oriented repetitive training suggests that biofeedback therapy should be delivered during functionally related dynamic movement to optimize motor function improvement. Current, advanced technologies facilitate the design of novel biofeedback systems that possess diverse parameters, advanced cue display, and sophisticated control systems for use in task-oriented biofeedback. In light of these advancements, this article: (1) reviews early biofeedback studies and their conclusions; (2) presents recent developments in biofeedback technologies and their applications to task-oriented biofeedback interventions; and (3) discusses considerations regarding the therapeutic system design and the clinical application of task-oriented biofeedback therapy. This review should provide a framework to further broaden the application of task-oriented biofeedback therapy in neuromotor rehabilitation.}, journal={Journal of NeuroEngineering and Rehabilitation}, publisher={Springer Science and Business Media LLC}, author={Huang, He and Wolf, Steven L and He, Jiping}, year={2006}, pages={11} }
 @inproceedings{chen_huang_xu_wallis_sundaram_rikakis_ingalls_olson_he_2006, title={The design of a real-time, multimodal biofeedback system for stroke patient rehabilitation}, ISBN={1595934472}, url={http://dx.doi.org/10.1145/1180639.1180804}, DOI={10.1145/1180639.1180804}, abstractNote={This paper presents a novel real-time, multi-modal biofeedback system for stroke patient therapy. The problem is important as traditional mechanisms of rehabilitation are monotonous, and do not incorporate detailed quantitative assessment of recovery in addition to traditional clinical schemes. We have been working on developing an experiential media system that integrates task dependent physical therapy and cognitive stimuli within an interactive, multimodal environment. The environment provides a purposeful, engaging, visual and auditory scene in which patients can practice functional therapeutic reaching tasks, while receiving different types of simultaneous feedback indicating measures of both performance and results. There are three contributions of this paper - (a) identification of features and goals for the functional task (b) The development of sophisticated feedback (auditory and visual) mechanisms that match the semantics of action of the task. We additionally develop novel action-feedback coupling mechanisms. (c) New metrics to validate the ability of the system to promote learnability, stylization and engagement. We have validated the system for nine subjects with excellent results.}, booktitle={Proceedings of the 14th annual ACM international conference on Multimedia  - MULTIMEDIA '06}, publisher={ACM Press}, author={Chen, Yinpeng and Huang, He and Xu, Weiwei and Wallis, Richard Isaac and Sundaram, Hari and Rikakis, Thanassis and Ingalls, Todd and Olson, Loren and He, Jiping}, year={2006}, pages={763–772} }
 @inproceedings{he_koeneman_schultz_huang_wanberg_herring_sugar_herman_koeneman_2005, title={Design of a Robotic Upper Extremity Repetitive Therapy Device}, volume={2005}, ISBN={0780390032}, url={http://dx.doi.org/10.1109/icorr.2005.1501060}, DOI={10.1109/icorr.2005.1501060}, abstractNote={Intensive repetitive therapy shows promise to improve motor function and quality of life for stroke patients. Intense therapies provided by individualized interaction between the patient and rehabilitation specialist to overcome upper extremity impairment are beneficial, however, they are expensive and difficult to evaluate quantitatively and objectively. The development of a pneumatic muscle (PM) driven therapeutic device, the RUPERT/spl trade/ has the potential of providing a low cost and safe take-home method of supplementing therapy in addition to in the clinic treatment. The device can also provide real-time, objective assessment of functional improvement from the therapy.}, booktitle={9th International Conference on Rehabilitation Robotics, 2005. ICORR 2005.}, publisher={IEEE}, author={He, J. and Koeneman, E.J. and Schultz, R.S. and Huang, H. and Wanberg, J. and Herring, D.E. and Sugar, T. and Herman, R. and Koeneman, J.B.}, year={2005}, month={Aug}, pages={95–98} }
 @inproceedings{huang_ingalls_olson_ganley_rikakis_he_2005, title={Interactive Multimodal Biofeedback for Task-Oriented Neural Rehabilitation}, volume={7 VOLS}, ISBN={0780387414}, url={http://dx.doi.org/10.1109/iembs.2005.1616988}, DOI={10.1109/iembs.2005.1616988}, abstractNote={Previous studies have suggested that task-oriented biofeedback training may be effective for functional motor improvement. The purpose of this project was to design an interactive, multimodal biofeedback system for the task-oriented training of goal-directed reaching. The central controller, based on a user context model, identifies the state of task performance using multisensing data and provides augmented feedback, through interactive 3D graphics and music, to encourage the patients' self-regulation and performance of the task. The design allows stroke patients to train with functional tasks, and receive real-time performance evaluation through successful processing of multimodal sensory feedback. In addition, the environment and training task is customizable. Overall, the system delivers an engaging training experience. Preliminary results of a pilot study involving stroke patients demonstrate the potential of the system to improve patients' reaching performance}, booktitle={2005 IEEE Engineering in Medicine and Biology 27th Annual Conference}, publisher={IEEE}, author={Huang, He and Ingalls, T. and Olson, L. and Ganley, K. and Rikakis, T. and He, Jiping}, year={2005}, pages={2547–2550} }
 @inproceedings{he_koeneman_schultz_herring_wanberg_huang_sugar_herman_koeneman_2005, title={RUPERT: A device for robotic upper extremity repetitive therapy}, volume={7 VOLS}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-33846912544&partnerID=MN8TOARS}, booktitle={Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings}, author={He, J. and Koeneman, E.J. and Schultz, R.S. and Herring, D.E. and Wanberg, J. and Huang, H. and Sugar, T. and Herman, R. and Koeneman, J.B.}, year={2005}, pages={6844–6847} }
 @inproceedings{he_koeneman_schultz_herring_wanberg_huang_sugar_herman_koeneman_2005, title={RUPERT: a Device for Robotic Upper Extremity Repetitive Therapy}, ISBN={0780387414}, url={http://dx.doi.org/10.1109/iembs.2005.1616077}, DOI={10.1109/iembs.2005.1616077}, abstractNote={We report the development and initial evaluation of a device for robotic assisted upper extremity repetitive therapy (RUPERTtrade). Intense repetitive physical therapies provided by individualized interaction between the patient and a rehabilitation specialist to overcome upper extremity impairment after stroke are beneficial, however, they are expensive and difficult to evaluate quantitatively and objectively. The need is urgent and growing for a low cost, safe and easy to use robotic device to assist the patient and the therapist to fully achieve the potential benefit of task-based repetitive physical therapies. We designed a pneumatic muscle (PM) driven therapeutic device, the RUPERTtrade, that is wearable and provides assistive forces required to move the arm during performance of several critical tasks of daily living. The robot has four degrees of freedom at shoulder, elbow and wrist. The sensors feedback position and force information for quantitative evaluation of task performance. It has the potential of providing a take-home method of supplementing therapy. The device can also provide real-time, objective assessment of functional improvement of therapy}, booktitle={2005 IEEE Engineering in Medicine and Biology 27th Annual Conference}, publisher={IEEE}, author={He, Jiping and Koeneman, E.J. and Schultz, R.S. and Herring, D.E. and Wanberg, J. and Huang, H. and Sugar, T. and Herman, R. and Koeneman, J.B.}, year={2005} }
 @inproceedings{huang_papandreou-suppappola_he_2004, title={Analysis of surface electromyogram during gait by modified matching pursuit decomposition}, ISBN={0780377893}, url={http://dx.doi.org/10.1109/iembs.2003.1280400}, DOI={10.1109/iembs.2003.1280400}, abstractNote={The matching pursuit decomposition (MPD) with a redundant Gabor dictionary (GMPD) is a popular method to identify the time-frequency content of nonstationary biomedical signals. However, it is time-consuming and suffers from the low decomposition efficiency when dealing with a signal with multiple nonlinear time-frequency structures (TFS) such as surface electromyograms (SEMG) recorded during gait. The modified MPD (MMPD) solves these problems by modifying the dictionary using elements with similar TFS as the signal to be decomposed. The reassigned spectrogram is used to obtain the TFS of the signal. This prior knowledge about the TFS of the signal is the key for the successful application of the MMPD. The MMPD combined with the reassigned spectrogram is applied in gait SEMG to compare the decomposition efficiency with GMPD, and to quantify the effects of the epidural spinal cord stimulation (ESCS) on the muscle activities of an incomplete spinal cord injured (ISCI) patient in this study. The results show that the MMPD combined with reassigned spectrogram can decompose SEMG with much higher efficiency than the GMPD. Moreover, the MMPD provides a quantitative measure to compare the difference of the time and frequency content between the gait SEMG with and without stimulation. These differences demonstrate that ESCS changes the timing pattern of SEMG during the gait, but does not significantly alter the centroid of the frequency content of SEMG.}, booktitle={Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (IEEE Cat. No.03CH37439)}, publisher={IEEE}, author={Huang, He and Papandreou-Suppappola, A. and He, Jiping}, year={2004}, month={Jun} }
 @inproceedings{carhart_willis_thompson_huang_d'luzansky_thresher_herman_he_2004, title={Mechanical and metabolic changes in gait performance with spinal cord stimulation and reflex-FES}, ISBN={0780377893}, url={http://dx.doi.org/10.1109/iembs.2003.1279657}, DOI={10.1109/iembs.2003.1279657}, abstractNote={In a previous study, we demonstrated that the application of lumbar epidural spinal cord stimulation (ESCS) produced marked improvements in walking performance of a tetraplegic with incomplete spinal cord injury (ISCI). The present study seeks to extend these results, as well as to compare the enhancement in walking performance provided by ESCS with that offered by reflex functional electrical stimulation (FES). A 48 year old ISCI participant (T8, ASIA C, 8 years post-injury) was provided with several months of partial weight bearing therapy (PWBT) with and without FES, followed by over-ground training assisted by ESCS and FES. Over-ground walking performance and metabolic response was subsequently evaluated under four conditions: no stimulation (NS), FES, ESCS, and ESCS+FES. Performance measures included: gait kinematics, average walking speed, maximum walking distance, pulmonary gas exchange, and the reliance on assistive devices. Stimulation of any type markedly improved locomotion, and reduced the O/sub 2/ cost of transport. FES was associated with a dramatic improvement in limb motion unilaterally, enhancing limb swing, and step length; walking speed and endurance were improved by factors of 2 and 4, respectively, while the O/sub 2/ cost of transport was reduced by 45% versus the NS condition. Despite less significant improvements in movement kinematics, ESCS resulted in further improvements in walking speed and endurance, a 57% reduction in the O/sub 2/ cost of transport, and a reduction in the reliance on the instrumented walker for body weight support. ESCS resulted in a respiratory exchange ratio (RER) suggesting marked reliance on fat oxidation for energy, similar to able-bodied walking at preferred speed. In contrast, FES elicited a CO/sub 2/ production consistent with carbohydrate dependence roughly similar to the NS condition. Superimposing FES on ESCS improved walking speed and endurance while reducing the participant's reliance on the walker for support, but apparently at the expense of greater carbohydrate dependence. We conclude that stimulation of the peripheral and central nervous system can facilitate walking in individuals with ISCI. Further, ESCS may facilitate a neural activation pattern that favors fat metabolism in a manner which more-closely resembles that operating in the able-bodied population.}, booktitle={Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (IEEE Cat. No.03CH37439)}, publisher={IEEE}, author={Carhart, M. and Willis, W. and Thompson, A. and Huang, H. and D'Luzansky, S. and Thresher, J. and Herman, R. and He, J.}, year={2004}, month={Jun} }
 @inproceedings{huang_he_2004, title={Utilization of biomechanical modeling in design of Robotic arm for rehabilitation of stroke patients}, volume={26 IV}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-11144273973&partnerID=MN8TOARS}, booktitle={Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings}, author={Huang, H. and He, J.}, year={2004}, pages={2718–2721} }
 @inproceedings{huang_jiping_2004, title={Utilization of biomechanical modeling in design of robotic arm for rehabilitation of stroke patients}, volume={4}, booktitle={Conference Proceedings of the IEEE Engineering in Medicine and Biology Society}, author={Huang, H. and Jiping, H.}, year={2004}, pages={2718–2721} }
 @inproceedings{huang_papandreou-suppappola_he_2003, title={Analysis of surface electromyogram during gait by modified matching pursuit decomposition}, volume={3}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-1542270968&partnerID=MN8TOARS}, booktitle={Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings}, author={Huang, H. and Papandreou-Suppappola, A. and He, J.}, year={2003}, pages={2402–2405} }
 @inproceedings{huang_he_carhart_d'luzansky_herman_2003, title={Change of muscle activation pattern by epidural stimulation on a SCI patient}, ISBN={0780376129}, url={http://dx.doi.org/10.1109/iembs.2002.1134412}, DOI={10.1109/iembs.2002.1134412}, abstractNote={Surface electromyograms (sEMG) of major leg muscles, recorded during treadmill walking, were analyzed to quantify the improvement in functional ambulation facilitated by epidural spinal cord stimulation (ESCS) compared with partial weight-bearing therapy (PWBT) alone. Due to the nonstationary and stochastic properties of the signal, advanced algorithms were developed to analyze sEMG recorded during this quasi-cyclic dynamic movement. Principal component analysis (PCA) and Cohen class time-frequency spectral analysis distinguished changes in the pattern of sEMG due to ESCS. Moreover, the results from both analysis techniques demonstrate ESCS improved muscle activation patterns during treadmill walking beyond that afforded by PWBT alone, providing an improved temporal match to normal EMG patterns.}, booktitle={Proceedings of the Second Joint 24th Annual Conference and the Annual Fall Meeting of the Biomedical Engineering Society] [Engineering in Medicine and Biology}, publisher={IEEE}, author={Huang, H. and He, J. and Carhart, M. and D'Luzansky, S. and Herman, R.}, year={2003}, month={Jun} }
 @inproceedings{kuchi_hiremagalur_huang_carhart_he_panchanathan_2003, title={DRAG: a database for recognition and analasys of gait}, volume={5242}, url={http://dx.doi.org/10.1117/12.515732}, DOI={10.1117/12.515732}, abstractNote={A novel approach is proposed for creating a standardized and comprehensive database for gait analysis. The field of gait analysis is gaining increasing attention for applications such as visual surveillance, human-computer interfaces, and gait recognition and rehabilitation. Numerous algorithms have been developed for analyzing and processing gait data; however, a standard database for their systematic evaluation does not exist. Instead, existing gait databases consist of subsets of kinematic, kinetic, and electromyographic activity recordings by different investigators, at separate laboratories, and under varying conditions. Thus, the existing databases are neither homogenous nor sufficiently populated to statistically validate the algorithms. In this paper, a methodology for creating a database is presented, which can be used as a common ground to test the performance of algorithms that rely upon external marker data, ground reaction loading data, and/or video images. The database consists of: (1) synchronized motion-capture data (3D marker data) obtained using external markers, (2) computed joint angles, and (3) ground reaction loading acquired with plantar pressure insoles. This database could be easily expanded to include synchronized video, which will facilitate further development of video-based algorithms for motion tracking. This eventually could lead to the realization of markerless gait tracking. Such a system would have extensive applications in gait recognition, as well as gait rehabilitation. The entire database (marker, angle, and force data) will be placed in the public domain, and made available for downloads over the World Wide Web.}, booktitle={Internet Multimedia Management Systems IV}, publisher={SPIE}, author={Kuchi, Prem and Hiremagalur, Raghu Ram V. and Huang, Helen and Carhart, Michael and He, Jiping and Panchanathan, Sethuraman}, editor={Smith, John R. and Panchanathan, Sethuraman and Zhang, TongEditors}, year={2003}, month={Nov}, pages={115–124} }
 @inproceedings{carhart_willis_thompson_huang_d’luzansky_thresher_herman_he_2003, title={Mechanical and metabolic changes in gait permormance with spinal cord stimulation and reflex-FES}, volume={2}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-1542361474&partnerID=MN8TOARS}, booktitle={Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings}, author={Carhart, M. and Willis, W. and Thompson, A. and Huang, H. and D’Luzansky, S. and Thresher, J. and Herman, R. and He, J.}, year={2003}, pages={1558–1561} }
 @inproceedings{huang_he_carhart_d’luzansky_herman_2002, title={Change of muscle activation pattern by epidural stimulation on a SCI patient}, volume={1}, url={http://www.scopus.com/inward/record.url?eid=2-s2.0-0036912655&partnerID=MN8TOARS}, booktitle={Annual International Conference of the IEEE Engineering in Medicine and Biology - Proceedings}, author={Huang, H. and He, J. and Carhart, M. and D’Luzansky, S. and Herman, R.}, year={2002}, pages={114–115} }