@article{su_jung_lu_wang_qing_xu_2024, title={Exploring the impact of human-robot interaction on workers' mental stress in collaborative assembly tasks}, volume={116}, ISSN={["1872-9126"]}, url={https://doi.org/10.1016/j.apergo.2024.104224}, DOI={10.1016/j.apergo.2024.104224}, abstractNote={Advances in robotics have contributed to the prevalence of human-robot collaboration (HRC). However, working and interacting with collaborative robots in close proximity can be psychologically stressful. Therefore, understanding the impacts of human-robot interaction (HRI) on mental stress is crucial for enhancing workplace well-being. To this end, this study investigated how the HRI factors – presence, complexity, and modality – affect the psychological stress of workers. We employed both the NASA-Task Load Index for subjective assessment and physiological metrics including galvanic skin responses, electromyography, and heart rate for objective evaluation. An experimental setup was implemented in which human operators worked together with a collaborative robot on Lego assembly tasks, using different interaction paradigms including pressing buttons, showing hand gestures, and giving verbal commands. The results revealed that the introduction of interactions during HRC helped reduce mental stress and that complex interactions resulted in higher mental stress than simple interactions. Meanwhile, using hand gestures led to significantly higher mental stress than pressing buttons and verbal commands. The findings provided practical insights for mitigating mental stress in the workplace and promoting wellness in the era of HRC.}, journal={APPLIED ERGONOMICS}, author={Su, Bingyi and Jung, SeHee and Lu, Lu and Wang, Hanwen and Qing, Liwei and Xu, Xu}, year={2024}, month={Apr} }
 @article{lu_xie_wang_su_jung_xu_2024, title={Factors Affecting Workers' Mental Stress in Handover Activities During Human-Robot Collaboration}, volume={1}, ISSN={["1547-8181"]}, url={https://doi.org/10.1177/00187208241226823}, DOI={10.1177/00187208241226823}, abstractNote={Objective This study investigated the effects of different approach directions, movement speeds, and trajectories of a co-robot’s end-effector on workers’ mental stress during handover tasks. }, journal={HUMAN FACTORS}, author={Lu, Lu and Xie, Ziyang and Wang, Hanwen and Su, Bingyi and Jung, Sehee and Xu, Xu}, year={2024}, month={Jan} }
 @article{lim_hsiao_xu_2024, title={Human–Robot Collaboration in Occupational Settings : An Introduction to the Special Issue}, url={https://doi.org/10.1080/24725838.2023.2339620}, DOI={10.1080/24725838.2023.2339620}, journal={IISE Transactions on Occupational Ergonomics and Human Factors}, author={Lim, Sol and Hsiao, Hongwei and Xu, Xu}, year={2024}, month={Apr} }
 @article{xie_lu_wang_li_xu_2023, title={An Image-Based Human-Robot Collision Avoidance Scheme: A Proof of Concept}, volume={6}, ISSN={["2472-5846"]}, DOI={10.1080/24725838.2023.2222651}, abstractNote={Occupational Applications:In modern industrial plants, collisions between humans and robots pose a significant risk to occupational safety. To address this concern, we sought to devise a reliable system for human-robot collision avoidance system employing computer vision. This system enables proactive prevention of dangerous collisions between humans and robots. In contrast to previous approaches, we used a standard RGB camera, making implementation more convenient and cost-effective. Furthermore, the proposed method greatly extends the effective detection range compared to previous studies, thereby enhancing its utility for monitoring large-scale workplaces.}, journal={IISE TRANSACTIONS ON OCCUPATIONAL ERGONOMICS & HUMAN FACTORS}, author={Xie, Ziyang and Lu, Lu and Wang, Hanwen and Li, Li and Xu, Xu}, year={2023}, month={Jun} }
 @article{xie_lu_wang_su_liu_xu_2023, title={Improving Workers' Musculoskeletal Health During Human-Robot Collaboration Through Reinforcement Learning}, volume={5}, ISSN={["1547-8181"]}, url={https://doi.org/10.1177/00187208231177574}, DOI={10.1177/00187208231177574}, abstractNote={Objective This study aims to improve workers’ postures and thus reduce the risk of musculoskeletal disorders in human-robot collaboration by developing a novel model-free reinforcement learning method. }, journal={HUMAN FACTORS}, author={Xie, Ziyang and Lu, Lu and Wang, Hanwen and Su, Bingyi and Liu, Yunan and Xu, Xu}, year={2023}, month={May} }
 @article{liu_chang_li_xu_2022, title={A Simple Method to Optimally Select Upper-Limb Joint Angle Trajectories from Two Kinect Sensors during the Twisting Task for Posture Analysis}, volume={22}, ISSN={["1424-8220"]}, DOI={10.3390/s22197662}, abstractNote={A trunk-twisting posture is strongly associated with physical discomfort. Measurement of joint kinematics to assess physical exposure to injuries is important. However, using a single Kinect sensor to track the upper-limb joint angle trajectories during twisting tasks in the workplace is challenging due to sensor view occlusions. This study provides and validates a simple method to optimally select the upper-limb joint angle data from two Kinect sensors at different viewing angles during the twisting task, so the errors of trajectory estimation can be improved. Twelve healthy participants performed a rightward twisting task. The tracking errors of the upper-limb joint angle trajectories of two Kinect sensors during the twisting task were estimated based on concurrent data collected using a conventional motion tracking system. The error values were applied to generate the error trendlines of two Kinect sensors using third-order polynomial regressions. The intersections between two error trendlines were used to define the optimal data selection points for data integration. The finding indicates that integrating the outputs from two Kinect sensor datasets using the proposed method can be more robust than using a single sensor for upper-limb joint angle trajectory estimations during the twisting task.}, number={19}, journal={SENSORS}, author={Liu, Pin-Ling and Chang, Chien-Chi and Li, Li and Xu, Xu}, year={2022}, month={Oct} }
 @article{wang_xie_lu_su_jung_xu_2022, title={A mobile platform-based app to assist undergraduate learning of human kinematics in biomechanics courses}, volume={142}, ISSN={["1873-2380"]}, DOI={10.1016/j.jbiomech.2022.111243}, abstractNote={Whole-body biomechanics examines different physical characteristics of the human body movement by applying principles of Newtonian mechanics. Therefore, undergraduate biomechanics courses are highly demanding in mathematics and physics. While the inclusion of laboratory experiences can augment student comprehension of biomechanics concepts, the cost and the required expertise associated with experiment equipment can be a burden of offering laboratory sessions. In this study, we developed a mobile app to facilitate learning human kinematics in biomechanics curriculums. First, a mobile-based computer-vision algorithm that is based on Convolutional pose machine (CPM), MobileNet V2, and TensorFlow Lite framework is adopted to reconstruct 2D human poses from the images collected by a mobile device camera. Key joint locations are then applied to the human kinematics variable estimator for human kinematics analysis. Simultaneously, students can view various kinematics data for a selected joint or body segment in real-time through the user interface of the mobile device. The proposed app can serve as a potential instructional tool to assist in conducting human motion experiments in biomechanics courses.}, journal={JOURNAL OF BIOMECHANICS}, author={Wang, Hanwen and Xie, Ziyang and Lu, Lu and Su, Bingyi and Jung, Sehee and Xu, Xu}, year={2022}, month={Sep} }
 @misc{lu_xie_wang_li_xu_2022, title={Mental stress and safety awareness during human-robot collaboration - Review}, volume={105}, ISSN={["1872-9126"]}, url={https://doi.org/10.1016/j.apergo.2022.103832}, DOI={10.1016/j.apergo.2022.103832}, abstractNote={Human-robot collaboration (HRC) is an emerging research area that has gained tremendous attention in both academia and industry. Yet, the feature that humans and robots sharing the workplace has led to safety concerns. In particular, the mental stress or safety awareness of human teammates during HRC remains unclear but is also of great importance to workplace safety. In this manuscript, we reviewed twenty-five studies for understanding the relationships between HRC and workers' mental stress or safety awareness. Specifically, we aimed to understand: (1) robot-related factors that may affect human workers' mental stress or safety awareness, (2) a number of measurements that could be used to evaluate workers' mental stress in HRC, and (3) various methods for measuring safety awareness that had been adopted or could be applied in HRC. According to our literature review, robot-related factors including robot characteristics, social touching and trajectory have relationships with workers' mental stress or safety awareness. For the measurement of mental stress and safety awareness, each method mentioned has its validity and rationality. Additionally, a discussion related to the potential co-robot actions to lower mental stress or improve safety awareness as well as future implications were provided.}, journal={APPLIED ERGONOMICS}, publisher={Elsevier BV}, author={Lu, Lu and Xie, Ziyang and Wang, Hanwen and Li, Li and Xu, Xu}, year={2022}, month={Nov} }
 @article{xie_li_xu_2022, title={Real-Time Driving Distraction Recognition Through a Wrist-Mounted Accelerometer}, volume={64}, ISSN={["1547-8181"]}, url={https://doi.org/10.1177/0018720821995000}, DOI={10.1177/0018720821995000}, abstractNote={Objective We propose a method for recognizing driver distraction in real time using a wrist-worn inertial measurement unit (IMU). }, number={8}, journal={HUMAN FACTORS}, publisher={SAGE Publications}, author={Xie, Ziyang and Li, Li and Xu, Xu}, year={2022}, month={Dec}, pages={1412–1428} }
 @article{wang_xie_lu_li_xu_2021, title={A computer-vision method to estimate joint angles and L5/S1 moments during lifting tasks through a single camera}, volume={129}, ISSN={["1873-2380"]}, DOI={10.1016/j.jbiomech.2021.110860}, abstractNote={Weight lifting is a risk factor of work-related low-back musculoskeletal disorders (MSD). From the ergonomics perspective, it is important to measure workers' body motion during a lifting task and estimate low-back joint moments to ensure the low-back biomechanical loadings are within the failure tolerance. With the recent development of advanced deep neural networks, an increasing number of computer vision algorithms have been presented to estimate 3D human poses through videos. In this study, we first performed a 3D pose estimation of lifting tasks using a single RGB camera and VideoPose3D, an open-source library with a fully convolutional model. Joint angle trajectories and L5/S1 joint moment were then calculated following a top-down inverse dynamic biomechanical model. To evaluate the accuracy of the computer-vision-based angular trajectories and L5/S1 joint moments, we conducted an experiment in which participants performed a variety of lifting tasks. The body motions of the participants were concurrently captured by an RGB camera and a laboratory-grade motion tracking system. The body joint angles and L5/S1 joint moments obtained from the camera were compared with those obtained from the motion tracking system. The results showed a strong correlation (r > 0.9, RMSE < 10°) between the two methods for shoulder flexion, trunk flexion, trunk rotation, and elbow flexion. The computer-vision-based method also yielded a good estimate for the total L5/S1 moment and the L5/S1 moment in the sagittal plane (r > 0.9, RMSE < 20 N·m). This study showed computer vision could facilitate safety practitioners to quickly identify the jobs with high MSD risks through field survey videos.}, journal={JOURNAL OF BIOMECHANICS}, author={Wang, Hanwen and Xie, Ziyang and Lu, Lu and Li, Li and Xu, Xu}, year={2021}, month={Dec} }
 @article{li_prabhu_xie_wang_lu_xu_2021, title={Lifting Posture Prediction With Generative Models for Improving Occupational Safety}, volume={51}, ISSN={["2168-2305"]}, url={https://doi.org/10.1109/THMS.2021.3102511}, DOI={10.1109/THMS.2021.3102511}, abstractNote={Lifting tasks have been identified to be highly associated with work-related low back pain. Posture prediction can be used for simulating workers’ posture of lifting tasks and thus facilitate the prevention of low back pain (LBP). This study adopts two generative models, conditional variational encoder and conditional generative adversarial network, to predict lifting postures. A regular feed-forward neural network (FNN) developed upon previous studies is also investigated for comparison purposes. Ground-truth lifting posture data collected by a motion capture system is used for training and testing the models. The models are trained with datasets of different size and loss functions, and the results are compared. The conditional variational autoencoder and the regular FNN achieved comparable top performance in lifting posture prediction in terms of accuracy and posture validity. Both generative models are able to partially capture the variability of constrained postures. Overall, the results prove that using a generative model is able to predict postures with reasonable accuracy and validity (RMSE of coordinates = 0.049 m; RMSE of joint angles = 19.58$^\circ$). The predicted postures can support biomechanical analysis and ergonomics assessment of a lifting task to reduce the risk of low back injuries.}, number={5}, journal={IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Li, Li and Prabhu, Saiesh and Xie, Ziyang and Wang, Hanwen and Lu, Lu and Xu, Xu}, year={2021}, month={Oct}, pages={494–503} }
 @article{li_martin_xu_2020, title={A novel vision-based real-time method for evaluating postural risk factors associated with musculoskeletal disorders}, volume={87}, ISSN={["1872-9126"]}, DOI={10.1016/j.apergo.2020.103138}, abstractNote={Real-time risk assessment for work-related musculoskeletal disorders (MSD) has been a challenging research problem. Previous methods such as using depth cameras suffered from limited visual range and wearable sensors could cause intrusiveness to the workers, both of which are less feasible for long-run on-site applications. This document examines a novel end-to-end implementation of a deep learning-based algorithm for rapid upper limb assessment (RULA). The algorithm takes normal RGB images as input and outputs the RULA action level, which is a further division of RULA grand score. Lifting postures collected in laboratory and posture data from Human 3.6 (a public human pose dataset) were used for training and evaluating the algorithm. Overall, the algorithm achieved 93% accuracy and 29 frames per second efficiency for detecting the RULA action level. The results also indicate that using data augmentation (a strategy to diversify the training data) can significantly improve the robustness of the model. The proposed method demonstrates its high potential for real-time on-site risk assessment for the prevention of work-related MSD. A demo video can be found at https://github.com/LLDavid/RULA_2DImage.}, journal={APPLIED ERGONOMICS}, author={Li, Li and Martin, Tara and Xu, Xu}, year={2020}, month={Sep} }
 @article{li_zhong_hutmacher_liang_horrey_xu_2020, title={Detection of driver manual distraction via image-based hand and ear recognition}, volume={137}, ISSN={["1879-2057"]}, DOI={10.1016/j.aap.2020.105432}, abstractNote={Driving distraction is a leading cause of fatal car accidents, and almost nine people are killed in the US each day because of distracting activities. Therefore, reducing the number of distraction-affected traffic accidents remains an imperative issue. A novel algorithm for detection of drivers’ manual distraction was proposed in this manuscript. The detection algorithm consists of two modules. The first module predicts the bounding boxes of the driver's right hand and right ear from RGB images. The second module takes the bounding boxes as input and predicts the type of distraction. 106,677 frames extracted from videos, which were collected from twenty participants in a driving simulator, were used for training (50%) and testing (50%). For distraction classification, the results indicated that the proposed framework could detect normal driving, using the touchscreen, and talking with a phone with F1-score 0.84, 0.69, 0.82, respectively. For overall distraction detection, it achieved F1-score of 0.74. The whole framework ran at 28 frames per second. The algorithm achieved comparable overall accuracy with similar research, and was more efficient than other methods. A demo video for the algorithm can be found at https://youtu.be/NKclK1bHRd4.}, journal={ACCIDENT ANALYSIS AND PREVENTION}, author={Li, Li and Zhong, Boxuan and Hutmacher, Clayton, Jr. and Liang, Yulan and Horrey, William J. and Xu, Xu}, year={2020}, month={Mar} }
 @article{li_xie_xu_2020, title={MOPED25: A multimodal dataset of full-body pose and motion in occupational tasks}, volume={113}, ISSN={["1873-2380"]}, DOI={10.1016/j.jbiomech.2020.110086}, abstractNote={In recent years, there has been a trend of using images and deep neural network-based computer vision algorithms to perform postural evaluation in workplace safety and ergonomics community. The performance of the computer vision algorithms, however, heavily relies on the generalizability of the posture dataset that was used for algorithm training. Current open-access posture datasets from the computer vision community mainly focus on the pose and motion of daily activities and lack the context in workplaces. In this study, a new posture dataset named, MOPED25 (Multimodal Occupational Posture Dataset with 25 tasks) is presented. This dataset includes full-body kinematics data and the synchronized videos of 11 participants, performing commonly seen tasks at workplaces. All the data has been made publicly available online. This dataset can serve as a benchmark for developing more robust computer vision algorithms for postural evaluation at workplaces.}, journal={JOURNAL OF BIOMECHANICS}, author={Li, Li and Xie, Ziyang and Xu, Xu}, year={2020}, month={Dec} }
 @article{mehrizi_peng_xu_zhang_li_2019, title={A Deep Neural Network-based method for estimation of 3D lifting motions}, volume={84}, ISSN={["1873-2380"]}, DOI={10.1016/j.jbiomech.2018.12.022}, abstractNote={The aim of this study is developing and validating a Deep Neural Network (DNN) based method for 3D pose estimation during lifting. The proposed DNN based method addresses problems associated with marker-based motion capture systems like excessive preparation time, movement obstruction, and controlled environment requirement. Twelve healthy adults participated in a protocol and performed nine lifting tasks with different vertical heights and asymmetry angles. They lifted a crate and placed it on a shelf while being filmed by two camcorders and a synchronized motion capture system, which directly measured their body movement. A DNN with two-stage cascaded structure was designed to estimate subjects’ 3D body pose from images captured by camcorders. Our DNN augmented Hourglass network for monocular 2D pose estimation with a novel 3D pose generator subnetwork, which synthesized information from all available views to predict accurate 3D pose. We validated the results against the marker-based motion capture system as a reference and examined the method performance under different lifting conditions. The average Euclidean distance between the estimated 3D pose and reference (3D pose error) on the whole dataset was 14.72 ± 2.96 mm. Repeated measures ANOVAs showed lifting conditions can affect the method performance e.g. 60° asymmetry angle and shoulder height lifting showed higher 3D pose error compare to other lifting conditions. The results demonstrated the capability of the proposed method for 3D pose estimation with high accuracy and without limitations of marker-based motion capture systems. The proposed method may be utilized as an on-site biomechanical analysis tool.}, journal={JOURNAL OF BIOMECHANICS}, author={Mehrizi, Rahil and Peng, Xi and Xu, Xu and Zhang, Shaoting and Li, Kang}, year={2019}, month={Feb}, pages={87–93} }
 @article{white_kaber_deng_xu_2019, title={Design Process for an Ergonomic Solution to the Police Duty Belt}, volume={789}, ISBN={["978-3-319-94483-8"]}, ISSN={["2194-5365"]}, DOI={10.1007/978-3-319-94484-5_1}, abstractNote={Police officers carry various devices on their duty belts for use during patrols. The weight of a loaded belt can range from ~25–35 lbs. Such loading can lead to overexertion and associated injuries (e.g., low back pain) as well as reduced officer performance leading to injuries from violence. In addition, the distribution of the load can compromise officer balance, leading to slips and falls. The objective of this research was to identify design issues with current duty belts and to develop a design framework for police department use in creating custom ergonomic configurations of equipment on a belt. The study was divided into three phases: a literature review, a field study, and design of the ergonomic belt configuration. The resulting design framework for duty belt equipment configuration may serve as a design guideline for police departments and may reduce the incidence of officer musculoskeletal injuries.}, journal={ADVANCES IN PHYSICAL ERGONOMICS & HUMAN FACTORS}, author={White, Melissa Mae and Kaber, David B. and Deng, Yulin and Xu, Xu}, year={2019}, pages={3–15} }
 @article{lin_xu_2019, title={Occupational cranking operations: The scapula perspective}, volume={75}, ISSN={["1872-9126"]}, DOI={10.1016/j.apergo.2018.09.011}, abstractNote={Cranking the landing gear is a common task performed by truck drivers to raise or lower trailers. This task poses a risk to the shoulder joint due to the required forceful exertion and the posture constrained to the hand-handle interface. As a potential occupational risk, there has been no definitive guideline for best practices among truck drivers. An operator can crank perpendicular (frontal) or parallel (sagittal) to the crank rotation. In this laboratory study, the effects of cranking method and resistance on scapular range of motion and shoulder muscle activity were observed in 12 participants. Scapular posture was measured using an optical motion tracking system. EMG was monitored on 16 muscles contributing to shoulder movement. The results show that during frontal cranking, the scapular range of protraction was 28 ± 11.6°, which was more than the sagittal cranking (23 ± 10.4°), indicating a decreased subacromial space and elevated shoulder impingement risk. Seven muscles (all three deltoid muscles, middle trapezius, supraspinatus, infraspinatus, and teres minor) demonstrated that when the crank resistance was low, the front cranking method resulted in lower activity than the side cranking. When the crank resistance was 20 Nm, the muscle activity on these seven muscles was greater when cranking from the front than from the side. Based on these observations, we suggest that when the resistance is low (lowering the trailer) the driver should stand facing the trailer. On the contrary, it is advantageous to stand parallel to the trailer and crank while raising the trailer to apply the full body strength to reduce the shoulder load.}, journal={APPLIED ERGONOMICS}, author={Lin, Jia-Hua and Xu, Xu}, year={2019}, month={Feb}, pages={129–133} }
 @article{mehrizi_peng_xu_xu_shaoting_2019, title={Predicting 3-D Lower Back Joint Load in Lifting: A Deep Pose Estimation Approach}, volume={49}, DOI={10.1109/THMS.2018.2884811}, abstractNote={Goal: Lifting is a common manual material handling task performed in the workplaces. It is considered as one of the main risk factors for work-related musculoskeletal disorders. An important criterion to identify the unsafe lifting task is the values of the net force and moment at L5/S1 joint. These values are mainly calculated in a laboratory environment, which utilizes marker-based sensors to collect three-dimensional (3-D) information and force plates to measure the external forces and moments. However, this method is usually expensive to set up, time-consuming in process, and sensitive to the surrounding environment. In this study, we propose a deep neural network (DNN)-based framework for 3-D pose estimation, which addresses the aforementioned limitations, and we employ the results for L5/S1 moment and force calculation. Methods: At the first step of the proposed framework, full body 3-D pose is captured using a DNN, then at the second step, estimated 3-D body pose along with the subject's anthropometric information is utilized to calculate L5/S1 join's kinetic by a top-down inverse dynamic algorithm. Results: To fully evaluate our approach, we conducted experiments using a lifting dataset consisting of 12 subjects performing various types of lifting tasks. The results are validated against a marker-based motion capture system as a reference. The grand mean ± SD of the total moment/force absolute errors across all the dataset was 9.06 ± 7.60 N·m/4.85 ± 4.85 N. Conclusion: The proposed method provides a reliable tool for assessment of the lower back kinetics during lifting and can be an alternative when the use of marker-based motion capture systems is not possible.}, number={1}, journal={IEEE TRANSACTIONS ON HUMAN-MACHINE SYSTEMS}, author={Mehrizi, Rahil and Peng, Metaxas and Xu, Dimitris N. and Xu, Zhang and Shaoting, Li}, year={2019}, pages={85–94} }
 @article{li_hutmacher_xu_2019, title={Video-Based Driver's Hand Tracking using Fast Normalized Cross Coefficient with Improved Computational Efficiency}, volume={2673}, ISSN={["2169-4052"]}, DOI={10.1177/0361198119841554}, abstractNote={ Driver distraction is one of the major causes for fatal car accidents. In a distracting activity, manual distraction is a triggered response of other types of distraction, such as cognitive and visual distraction. Therefore, recognition of manual distraction can contribute to the monitoring of overall drivers’ distraction. In this study, a computer vision-based method to track hand movement from the recorded driving behavior is proposed. This method integrates a low computational cost template matching algorithm using fast normalized cross coefficient (NCC) and a novel searching strategy. The proposed method was evaluated by the VIVA hand tracking data set. It achieves 50.83% of marginal accuracy percentage (mAP), 42.18% of multiple object tracking accuracy (MOTA), 31.56% of mostly tracked (MT), and 19.29% of mostly lost (ML), and it outperformed a state of the art algorithm in MOTA and MT. Additionally, the computational cost of the proposed method is greatly improved, and it can run at around 11.1 frames per second. The outcome of this research will further assist driving distraction recognition and mitigation, and improve driving safety. }, number={8}, journal={TRANSPORTATION RESEARCH RECORD}, author={Li, Li and Hutmacher, Clayton M., Jr. and Xu, Xu}, year={2019}, month={Aug}, pages={233–241} }
 @article{mehrizi_peng_xu_zhang_metaxas_li_2018, title={A computer vision based method for 3D posture estimation of symmetrical lifting}, volume={69}, ISSN={["1873-2380"]}, DOI={10.1016/j.jbiomech.2018.01.012}, abstractNote={Work-related musculoskeletal disorders (WMSD) are commonly observed among the workers involved in material handling tasks such as lifting. To improve work place safety, it is necessary to assess musculoskeletal and biomechanical risk exposures associated with these tasks. Such an assessment has been mainly conducted using surface marker-based methods, which is time consuming and tedious. During the past decade, computer vision based pose estimation techniques have gained an increasing interest and may be a viable alternative for surface marker-based human movement analysis. The aim of this study is to develop and validate a computer vision based marker-less motion capture method to assess 3D joint kinematics of lifting tasks. Twelve subjects performing three types of symmetrical lifting tasks were filmed from two views using optical cameras. The joints kinematics were calculated by the proposed computer vision based motion capture method as well as a surface marker-based motion capture method. The joint kinematics estimated from the computer vision based method were practically comparable to the joint kinematics obtained by the surface marker-based method. The mean and standard deviation of the difference between the joint angles estimated by the computer vision based method and these obtained by the surface marker-based method was 2.31 ± 4.00°. One potential application of the proposed computer vision based marker-less method is to noninvasively assess 3D joint kinematics of industrial tasks such as lifting.}, journal={JOURNAL OF BIOMECHANICS}, author={Mehrizi, Rahil and Peng, Xi and Xu, Xu and Zhang, Shaoting and Metaxas, Dimitris and Li, Kang}, year={2018}, month={Mar}, pages={40–46} }
 @article{chang_xu_2018, title={Identification of heel strike under a slippery condition}, volume={66}, ISSN={["1872-9126"]}, DOI={10.1016/j.apergo.2017.08.004}, abstractNote={Kinematics at heel strike instant (HSI) has been used to quantify slip severity. However, methods to identify HSI remain ambiguous and have not been evaluated under slippery conditions. A glass force plate was used to observe the contact interface between shoe and floor under slippery conditions. HSIs identified from the video captured beneath the force plate and from the force plate and kinematics were compared. The results showed that HSIs identified with the video were closer to those identified with the normal force threshold (NFT) (9.0 ms ± 5.5 ms) than were most of those identified with kinematics. Slips with a longer distance travelled between NFT HSI and video HSI had a larger heel horizontal velocity (>0.8 m/s) and a smaller foot angular velocity (<100deg/s) at the NFT instant, and were still part of the forward swing. The results show that improved methods are needed over NFT to identify HSI, especially under slippery conditions.}, journal={APPLIED ERGONOMICS}, author={Chang, Wen-Ruey and Xu, Xu}, year={2018}, month={Jan}, pages={32–40} }
 @article{yu_xu_lin_2018, title={Impact of posture choice on one-handed pull strength variations at low, waist, and overhead pulling heights}, volume={64}, ISSN={["1872-8219"]}, DOI={10.1016/j.ergon.2017.07.004}, abstractNote={Manual exertions with one hand are commonly performed; however, current one-handed normative strength databases note large strength variations not explained by participant demographics, e.g., age and gender. This study models how the postures that participants choose impact their one-handed pull strength. Lower-extremity posture, defined as distance from handle, were measured for 31 participants performing two maximum one-handed frontal pull exertions at three handle heights, resulting in a strength-posture dataset of 186 maximal exertions. Regression models with anthropometric and lower extremity positions were created for each handle height. Smaller distances between the front-foot and handle were associated with higher strength. Larger back-foot distance to handle was positively associated with strength for low and waist-level handle heights. Larger hip distances from the handle increased pull strengths and interacted with feet positions. This study quantifies the relationship between one-handed pull strength with anthropometry and postures and provides practitioners posture-based equations that can be used to estimate safe pull strengths given workplace posture constraints. Maximum one-handed pull strength models are presented that can assist practitioners in: 1) assessing whether strength demands are within worker's capability in the workplace, 2) designing workplaces that better accommodate the range of postures chosen during maximum exertion, and 3) identify workplace layout constraints that adversely impact strength capabilities.}, journal={INTERNATIONAL JOURNAL OF INDUSTRIAL ERGONOMICS}, author={Yu, Denny and Xu, Xu and Lin, Jia-Hua}, year={2018}, month={Mar}, pages={226–234} }
 @article{lin_kirlik_xu_2018, title={New technologies in human factors and ergonomics research and practice}, volume={66}, ISSN={["1872-9126"]}, DOI={10.1016/j.apergo.2017.08.012}, journal={APPLIED ERGONOMICS}, author={Lin, Jia-Hua and Kirlik, Alex and Xu, Xu}, year={2018}, month={Jan}, pages={179–181} }
 @article{mehrizi_peng_tang_xu_metaxas_li_2018, title={Toward Marker-free 3D Pose Estimation in Lifting: A Deep Multi-view Solution}, ISSN={["2326-5396"]}, DOI={10.1109/FG.2018.00078}, abstractNote={Lifting is a common manual material handling task performed in the workplaces. It is considered as one of the main risk factors for Work-related Musculoskeletal Disorders. To improve work place safety, it is necessary to assess musculoskeletal and biomechanical risk exposures associated with these tasks, which requires very accurate 3D pose. Existing approaches mainly utilize marker-based sensors to collect 3D information. However, these methods are usually expensive to setup, timeconsuming in process, and sensitive to the surrounding environment. In this study, we propose a multi-view based deep perceptron approach to address aforementioned limitations. Our approach consists of two modules: a "view-specific perceptron" network extracts rich information independently from the image of view, which includes both 2D shape and hierarchical texture information; while a "multi-view integration" network synthesizes information from all available views to predict accurate 3D pose. To fully evaluate our approach, we carried out comprehensive experiments to compare different variants of our design. The results prove that our approach achieves comparable performance with former marker-based methods, i.e. an average error of 14:72 ± 2:96 mm on the lifting dataset. The results are also compared with state-of-the-art methods on HumanEva- I dataset [1], which demonstrates the superior performance of our approach.}, journal={PROCEEDINGS 2018 13TH IEEE INTERNATIONAL CONFERENCE ON AUTOMATIC FACE & GESTURE RECOGNITION (FG 2018)}, author={Mehrizi, Rahil and Peng, Xi and Tang, Zhiqiang and Xu, Xu and Metaxas, Dimitris and Li, Kang}, year={2018}, pages={485–491} }
 @article{hu_kim_ning_xu_2018, title={Using a deep learning network to recognise low back pain in static standing}, volume={61}, ISSN={["1366-5847"]}, DOI={10.1080/00140139.2018.1481230}, abstractNote={Abstract Low back pain (LBP) remains one of the most prevalent musculoskeletal disorders, while algorithms that able to recognise LBP patients from healthy population using balance performance data are rarely seen. In this study, human balance and body sway performance during standing trials were utilised to recognise chronic LBP populations using deep neural networks. To be specific, 44 chronic LBP and healthy individuals performed static standing tasks, while their spine kinematics and centre of pressure were recorded. A deep learning network with long short-term memory units was used for training, prediction and implementation. The performance of the model was evaluated by: (a) overall accuracy, (b) precision, (c) recall, (d) F1 measure, (e) receiver-operating characteristic and (f) area under the curve. Results indicated that deep neural networks could recognise LBP populations with precision up to 97.2% and recall up to 97.2%. Meanwhile, the results showed that the model with the C7 sensor output performed the best. Practitioner summary: Low back pain (LBP) remains the most common musculoskeletal disorder. In this study, we investigated the feasibility of applying artificial intelligent deep neural network in detecting LBP population from healthy controls with their kinematics data. Results showed a deep learning network can solve the above classification problem with both promising precision and recall performance.}, number={10}, journal={ERGONOMICS}, author={Hu, Boyi and Kim, Chong and Ning, Xiaopeng and Xu, Xu}, year={2018}, pages={1374–1381} }
 @article{xu_lin_mcgorry_2017, title={An entropy-assisted musculoskeletal shoulder model}, volume={33}, ISSN={["1873-5711"]}, DOI={10.1016/j.jelekin.2017.01.010}, abstractNote={Optimization combined with a musculoskeletal shoulder model has been used to estimate mechanical loading of musculoskeletal elements around the shoulder. Traditionally, the objective function is to minimize the summation of the total activities of the muscles with forces, moments, and stability constraints. Such an objective function, however, tends to neglect the antagonist muscle co-contraction. In this study, an objective function including an entropy term is proposed to address muscle co-contractions. A musculoskeletal shoulder model is developed to apply the proposed objective function. To find the optimal weight for the entropy term, an experiment was conducted. In the experiment, participants generated various 3-D shoulder moments in six shoulder postures. The surface EMG of 8 shoulder muscles was measured and compared with the predicted muscle activities based on the proposed objective function using Bhattacharyya distance and concordance ratio under different weight of the entropy term. The results show that a small weight of the entropy term can improve the predictability of the model in terms of muscle activities. Such a result suggests that the concept of entropy could be helpful for further understanding the mechanism of muscle co-contractions as well as developing a shoulder biomechanical model with greater validity.}, journal={JOURNAL OF ELECTROMYOGRAPHY AND KINESIOLOGY}, author={Xu, Xu and Lin, Jia-hua and McGorry, Raymond W.}, year={2017}, month={Apr}, pages={103–110} }
 @article{pajoutan_xu_cavuoto_2017, title={The effect of obesity on postural stability during a standardized lifting task}, volume={14}, ISSN={1545-9624 1545-9632}, url={http://dx.doi.org/10.1080/15459624.2016.1237032}, DOI={10.1080/15459624.2016.1237032}, abstractNote={ABSTRACT The objective of this study was to assess the effect of obesity on postural stability during a standardized lifting task. Twelve young males, six obese and six non-obese, completed three replications of repeated six lifts (at a rate of six lifts per minutes) at two levels of loads (10% and 25% of capacity) crossed with two levels of orientation (0° and 45° from sagittal plane). Postural stability measures showed that center of pressure sway path and sway area were ∼21% and ∼53% lower with obesity, respectively. Additionally, frequency band of amplitude spectrum in the medial lateral direction at 0° lifting orientation was significantly lower with obesity. The results suggest that obesity, as measured by body mass index, does not impair balance control in healthy young males when lifting load is relative to the capacity.}, number={3}, journal={Journal of Occupational and Environmental Hygiene}, publisher={Informa UK Limited}, author={Pajoutan, Mojdeh and Xu, Xu and Cavuoto, Lora A.}, year={2017}, pages={180–186} }
 @article{mehrizi_xu_zhang_pavlovic_metaxas_li_2017, title={Using a marker-less method for estimating L5/S1 moments during symmetrical lifting}, volume={65}, ISSN={["1872-9126"]}, DOI={10.1016/j.apergo.2017.01.007}, abstractNote={The aim of this study is to analyze the validity of a computer vision-based method to estimate 3D L5/S1 joint moment during symmetrical lifting. An important criterion to identify the non-ergonomic lifting task is the value of net moment at L5/S1 joint. This is usually calculated in a laboratory environment which is not practical for on-site biomechanical analysis. The validity of the proposed method, was assessed externally by comparing the results with a lab-based reference method and internally by comparing the estimated L5/S1 joint moments from top-down model and bottom-up model. It was shown that no significant differences in peak and mean moments between the two methods and intra-class correlation coefficients revealed excellent reliability of the proposed method (>0.91). The proposed method provides a reliable tool for assessment of lower back loads during occupational lifting and can be an alternative when the use of marker-based motion tracking systems is not possible.}, journal={APPLIED ERGONOMICS}, author={Mehrizi, Rahil and Xu, Xu and Zhang, Shaoting and Pavlovic, Vladimir and Metaxas, Dimitris and Li, Kang}, year={2017}, month={Nov}, pages={541–550} }
 @article{xu_robertson_chen_lin_mcgorry_2017, title={Using the Microsoft Kinect (TM) to assess 3-D shoulder kinematics during computer use}, volume={65}, ISSN={["1872-9126"]}, DOI={10.1016/j.apergo.2017.04.004}, abstractNote={Shoulder joint kinematics has been used as a representative indicator to investigate musculoskeletal symptoms among computer users for office ergonomics studies. The traditional measurement of shoulder kinematics normally requires a laboratory-based motion tracking system which limits the field studies. In the current study, a portable, low cost, and marker-less Microsoft Kinect™ sensor was examined for its feasibility on shoulder kinematics measurement during computer tasks. Eleven healthy participants performed a standardized computer task, and their shoulder kinematics data were measured by a Kinect sensor and a motion tracking system concurrently. The results indicated that placing the Kinect sensor in front of the participants would yielded a more accurate shoulder kinematics measurements then placing the Kinect sensor 15° or 30° to one side. The results also showed that the Kinect sensor had a better estimate on shoulder flexion/extension, compared with shoulder adduction/abduction and shoulder axial rotation. The RMSE of front-placed Kinect sensor on shoulder flexion/extension was less than 10° for both the right and the left shoulder. The measurement error of the front-placed Kinect sensor on the shoulder adduction/abduction was approximately 10° to 15°, and the magnitude of error is proportional to the magnitude of that joint angle. After the calibration, the RMSE on shoulder adduction/abduction were less than 10° based on an independent dataset of 5 additional participants. For shoulder axial rotation, the RMSE of front-placed Kinect sensor ranged between approximately 15° to 30°. The results of the study suggest that the Kinect sensor can provide some insight on shoulder kinematics for improving office ergonomics.}, journal={APPLIED ERGONOMICS}, author={Xu, Xu and Robertson, Michelle and Chen, Karen B. and Lin, Jia-hua and McGorry, Raymond W.}, year={2017}, month={Nov}, pages={418–423} }
 @article{harris-adamson_mielke_xu_lin_2016, title={Ergonomic evaluation of standard and alternative pallet jack handless}, volume={54}, ISSN={["1872-8219"]}, DOI={10.1016/j.ergon.2016.05.004}, abstractNote={Transportation of materials using a pallet jack pulled behind the operator is common due to the visual advantages while transporting fully loaded pallets. The objective of this laboratory study was to quantify muscle activity, posture, and low back compressive and shear forces while completing typical pallet jack activities using a standard handle that required one handed pulling of a pallet jack compared to an alternative handle that allowed for two handed pushing. Participants (n = 14) performed six to ten trials of common pallet jack tasks (straight travel and turning) with each handle. Posture analysis of the trunk and right upper extremity was performed using Motion Analysis (Santa Rosa, CA, USA) and back compressive and shear forces were analyzed using 3D Static Strength Prediction Program (University of Michigan, Ann Arbor, MI). Activity of the upper trapezius (UT), pectoralis major (PM), flexor digitorum superficialis (FDS) and extensor digitorum (ED) muscles were recorded (Telemyo 2400 T, Noraxon, Scottsdale, Arizona) and normalized to percent reference voluntary contraction values. All outcomes were compared using the paired t-test. Peak and mean muscle activity of the PM (p < 0.001) and ED (p < 0.01) were significantly higher using the alternative push handle during all three tasks. There were larger compressive forces at L4/L5 (p < 0.08) and L5/S1 (p < 0.002) using the alternative handle, and greater shear forces using the standard handle at both L4/L5 (p < 0.0001) and L5/S1 (p < 0.000). The standard handle outperformed the alternative handle with regard to muscle activity. The alternative handle had significantly greater compressive forces at L5/S1 due to the pushing nature of the hand-handle interface, yet the standard handle increased shear forces at both L4/L5 and L5/S1 levels in the low back. In this analysis, there was not a clear benefit to using either handle in terms of trunk strength capacity and varied benefits and drawbacks to each handle when comparing compressive and shear forces in the low back. However, given favorable subjective reports described in a prior publication, and the increased reliance on dynamic versus passive force production, facilitating a workers' ability to push a pallet jack while travelling with large loads is worth further investigation.}, journal={INTERNATIONAL JOURNAL OF INDUSTRIAL ERGONOMICS}, author={Harris-Adamson, Carisa and Mielke, Alexis and Xu, Xu and Lin, Jia-Hua}, year={2016}, month={Jul}, pages={113–119} }
 @article{xu_dickerson_lin_mcgorry_2016, title={Evaluation of regression-based 3-D shoulder rhythms}, volume={29}, ISSN={["1873-5711"]}, DOI={10.1016/j.jelekin.2015.07.005}, abstractNote={The movements of the humerus, the clavicle, and the scapula are not completely independent. The coupled pattern of movement of these bones is called the shoulder rhythm. To date, multiple studies have focused on providing regression-based 3-D shoulder rhythms, in which the orientations of the clavicle and the scapula are estimated by the orientation of the humerus. In this study, six existing regression-based shoulder rhythms were evaluated by an independent dataset in terms of their predictability. The datasets include the measured orientations of the humerus, the clavicle, and the scapula of 14 participants over 118 different upper arm postures. The predicted orientations of the clavicle and the scapula were derived from applying those regression-based shoulder rhythms to the humerus orientation. The results indicated that none of those regression-based shoulder rhythms provides consistently more accurate results than the others. For all the joint angles and all the shoulder rhythms, the RMSE are all greater than 5°. Among those shoulder rhythms, the scapula lateral/medial rotation has the strongest correlation between the predicted and the measured angles, while the other thoracoclavicular and thoracoscapular bone orientation angles only showed a weak to moderate correlation. Since the regression-based shoulder rhythm has been adopted for shoulder biomechanical models to estimate shoulder muscle activities and structure loads, there needs to be further investigation on how the predicted error from the shoulder rhythm affects the output of the biomechanical model.}, journal={JOURNAL OF ELECTROMYOGRAPHY AND KINESIOLOGY}, author={Xu, Xu and Dickerson, Clark R. and Lin, Jia-hua and McGorry, Raymond W.}, year={2016}, month={Aug}, pages={28–33} }
 @article{catena_xu_2016, title={Hip and knee net joint moments that correlate with success in lateral load transfers over a low friction surface}, journal={Ergonomics}, author={Catena, R. D. and Xu, X.}, year={2016}, pages={1–9} }
 @article{xu_mcgorry_chou_lin_chang_2015, title={Accuracy of the Microsoft Kinect (TM) for measuring gait parameters during treadmill walking}, volume={42}, ISSN={["1879-2219"]}, DOI={10.1016/j.gaitpost.2015.05.002}, abstractNote={The measurement of gait parameters normally requires motion tracking systems combined with force plates, which limits the measurement to laboratory settings. In some recent studies, the possibility of using the portable, low cost, and marker-less Microsoft Kinect sensor to measure gait parameters on over-ground walking has been examined. The current study further examined the accuracy level of the Kinect sensor for assessment of various gait parameters during treadmill walking under different walking speeds. Twenty healthy participants walked on the treadmill and their full body kinematics data were measured by a Kinect sensor and a motion tracking system, concurrently. Spatiotemporal gait parameters and knee and hip joint angles were extracted from the two devices and were compared. The results showed that the accuracy levels when using the Kinect sensor varied across the gait parameters. Average heel strike frame errors were 0.18 and 0.30 frames for the right and left foot, respectively, while average toe off frame errors were -2.25 and -2.61 frames, respectively, across all participants and all walking speeds. The temporal gait parameters based purely on heel strike have less error than the temporal gait parameters based on toe off. The Kinect sensor can follow the trend of the joint trajectories for the knee and hip joints, though there was substantial error in magnitudes. The walking speed was also found to significantly affect the identified timing of toe off. The results of the study suggest that the Kinect sensor may be used as an alternative device to measure some gait parameters for treadmill walking, depending on the desired accuracy level.}, number={2}, journal={GAIT & POSTURE}, author={Xu, Xu and McGorry, Raymond W. and Chou, Li-Shan and Lin, Jia-hua and Chang, Chien-chi}, year={2015}, month={Jul}, pages={145–151} }
 @article{chiu_chang_dennerlein_xu_2015, title={Age-related differences in inter-joint coordination during stair walking transitions}, volume={42}, ISSN={["1879-2219"]}, DOI={10.1016/j.gaitpost.2015.05.003}, abstractNote={Stair negotiation is one of the most difficult and hazardous locomotor tasks for older adults with fall-related accidences reported frequently. Since knowledge about inter-joint coordination during stair walking provides insights to age-related changes in neuromuscular control of gait that can inform prevention or intervention strategies, the current study investigated the effect of age on the pattern and variability of inter-joint coordination during stair-floor transitions during gait. Gait and motion analyses of the lower extremities of 20 young and 20 older adults during floor to stair (F-S) and stair to floor (S-F) walking transitions provided continuous measures of relative phase (CRP) that assessed inter-joint coordination of the hip, knee, and angle joints. The mean absolute relative phase (MARP) and deviation phase (DP) provided descriptive metrics for CRP pattern and variability respectively. For hip-knee CRP pattern, older adults demonstrated significantly smaller MARP than young adults in stance and most swing phases during F-S and S-F. For knee-ankle, older adults showed a significant smaller MARP of the trailing limb during S-F than young adults. In most stance and swing phases, the hip-knee DP values of older adults were significantly lower than that of young adults. Significant lower knee-ankle DP values of older adults were only detected in swing phase during S-F. The findings suggest that normal aging adults have less independent control of adjacent joints compared to younger adults suggesting they have less flexibility to modulate inter-joints coordination appropriately during stair walking transitions.}, number={2}, journal={GAIT & POSTURE}, author={Chiu, Shiu-Ling and Chang, Chien-Chi and Dennerlein, Jack T. and Xu, Xu}, year={2015}, month={Jul}, pages={152–157} }
 @article{xu_lin._2015, title={Effects of Working Environment Factors and Operator Experience on Upper Extremity Mechanical Properties During Powered Hand Tool Use}, volume={3}, DOI={10.1080/21577323.2014.968693}, abstractNote={OCCUPATIONAL APPLICATIONS Powered hand tools are widely used during assembly operations. The hand displacement during powered hand-tool use has been identified as a potential risk factor for upper extremity musculoskeletal disorders. In the current study, the mechanical properties of the upper extremity were identified, which represent muscle capacity to react to an impulsive power tool torque loading and affect the responsive hand displacement. These properties were obtained from among experienced and inexperienced participants under various operating configurations, including working heights, horizontal working distances, tool moments of inertia, and joint type. The results indicated that operating configurations and experiences affected the mechanical properties of upper extremities in different ways. This research may help in future studies on powered hand-tool work-station design, for example by improving parameters in biomechanical models. TECHNICAL ABSTRACT Rationale: Hand-tool displacement during powered hand-tool use is a potential risk factor for upper extremity injuries and is correlated to the subjective discomfort level. The upper extremity has been modeled as a second-order linear system to describe the hand-tool response. While previous studies have found that working environment factors and operator experience significantly affect the hand-tool response during powered tool use, how those factors affect the mechanical properties of the upper extremity has not been investigated. Purpose: This study assessed the mechanical properties of the upper extremity under various working environment factors and operator experience levels. Method: A least-squares method was used to identify the mechanical properties of the upper extremity during powered hand-tool use, directly from the dynamics of hand-tool response. Results: Working heights, horizontal working distances, hand tool moments of inertia, joint type, and experience significantly affected some mechanical properties of the upper extremities in various operating configurations. In addition, stiffness and damping coefficients of the upper extremities were greater than those values identified from a free oscillation system in a previous study. Conclusions: Mechanical properties of the upper extremities can be used to predict hand displacement during powered hand-tool use. The current results provide additional information to improve the understanding of operator reactions to powered hand tools.}, number={2}, journal={IIE Transactions on Occupational Ergonomics and Human Factors}, author={Xu, Xu and Lin., J.}, year={2015}, pages={81–90} }
 @article{chen_xu_lin_radwin_2015, title={Evaluation of older driver head functional range of motion using portable immersive virtual reality}, volume={70}, ISSN={0531-5565}, url={http://dx.doi.org/10.1016/J.EXGER.2015.08.010}, DOI={10.1016/j.exger.2015.08.010}, abstractNote={The number of drivers over 65 years of age continues to increase. Although neck rotation range has been identified as a factor associated with self-reported crash history in older drivers, it was not consistently reported as indicators of older driver performance or crashes across previous studies. It is likely that drivers use neck and trunk rotation when driving, and therefore the functional range of motion (ROM) (i.e. overall rotation used during a task) of older drivers should be further examined.Evaluate older driver performance in an immersive virtual reality, simulated, dynamic driving blind spot target detection task.A cross-sectional laboratory study recruited twenty-six licensed drivers (14 young between 18 and 35 years, and 12 older between 65 to 75 years) from the local community. Participants were asked to detect targets by performing blind spot check movements while neck and trunk rotation was tracked. Functional ROM, target detection success, and time to detection were analyzed.In addition to neck rotation, older and younger drivers on average rotated their trunks 9.96° and 18.04°, respectively. The younger drivers generally demonstrated 15.6° greater functional ROM (p<.001), were nearly twice as successful in target detection due to target location (p=.008), and had 0.46 s less target detection time (p=.016) than the older drivers.Assessing older driver functional ROM may provide more comprehensive assessment of driving ability than neck ROM. Target detection success and time to detection may also be part of the aging process as these measures differed between driver groups.}, journal={Experimental Gerontology}, publisher={Elsevier BV}, author={Chen, Karen B. and Xu, Xu and Lin, Jia-Hua and Radwin, Robert G.}, year={2015}, month={Oct}, pages={150–156} }
 @article{catena_xu_2015, title={Lower extremity kinematics that correlate with success in lateral load transfers over a low friction surface}, volume={58}, ISSN={["1366-5847"]}, DOI={10.1080/00140139.2015.1016122}, abstractNote={We previously studied balance during lateral load transfers, but were left without explanation of why some individuals were successful in novel low friction conditions and others were not. Here, we retrospectively examined lower extremity kinematics between successful (SL) and unsuccessful (UL) groups to determine what characteristics may improve low friction performance. Success versus failure over a novel slippery surface was used to dichotomise 35 healthy working-age individuals into the two groups (SL and UL). Participants performed lateral load transfers over three sequential surface conditions: high friction, novel low friction, and practiced low friction. The UL group used a wide stance with rotation mostly at the hips during the high and novel low friction conditions. To successfully complete the practiced low friction task, they narrowed their stance and pivoted both feet and torso towards the direction of the load, similar to the SL group in all conditions. This successful kinematic method potentially results in reduced muscle demand throughout the task. Practitioner Summary: The reason for this paper is to retrospectively examine the different load transfer strategies that are used in a low friction lateral load transfer. We found stance width to be the major source of success, while sagittal plane motion was altered to potentially maintain balance.}, number={9}, journal={ERGONOMICS}, author={Catena, Robert D. and Xu, Xu}, year={2015}, pages={1571–1580} }
 @article{xu_chen_lin_radwin_2015, title={The accuracy of the Oculus Rift virtual reality head-mounted display during cervical spine mobility measurement}, volume={48}, ISSN={["1873-2380"]}, DOI={10.1016/j.jbiomech.2015.01.005}, abstractNote={An inertial sensor-embedded virtual reality (VR) head-mounted display, the Oculus Rift (the Rift), monitors head movement so the content displayed can be updated accordingly. While the Rift may have potential use in cervical spine biomechanics studies, its accuracy in terms of cervical spine mobility measurement has not yet been validated. In the current study, a VR environment was designed to guide participants to perform prescribed neck movements. The cervical spine kinematics was measured by both the Rift and a reference motion tracking system. Comparison of the kinematics data between the Rift and the tracking system indicated that the Rift can provide good estimates on full range of motion (from one side to the other side) during the performed task. Because of inertial sensor drifting, the unilateral range of motion (from one side to neutral posture) derived from the Rift is more erroneous. The root-mean-square errors over a 1-min task were within 10° for each rotation axis. The error analysis further indicated that the inertial sensor drifted approximately 6° at the beginning of a trial during the initialization. This needs to be addressed when using the Rift in order to more accurately measure cervical spine kinematics. It is suggested that the front cover of the Rift should be aligned against a vertical plane during its initialization.}, number={4}, journal={JOURNAL OF BIOMECHANICS}, author={Xu, Xu and Chen, Karen B. and Lin, Jia-Hua and Radwin, Robert G.}, year={2015}, month={Feb}, pages={721–724} }
 @article{xu_mcgorry_2015, title={The validity of the first and second generation Microsoft Kinect (TM) for identifying joint center locations during static postures}, volume={49}, ISSN={["1872-9126"]}, DOI={10.1016/j.apergo.2015.01.005}, abstractNote={The Kinect™ sensor released by Microsoft is a low-cost, portable, and marker-less motion tracking system for the video game industry. Since the first generation Kinect sensor was released in 2010, many studies have been conducted to examine the validity of this sensor when used to measure body movement in different research areas. In 2014, Microsoft released the computer-used second generation Kinect sensor with a better resolution for the depth sensor. However, very few studies have performed a direct comparison between all the Kinect sensor-identified joint center locations and their corresponding motion tracking system-identified counterparts, the result of which may provide some insight into the error of the Kinect-identified segment length, joint angles, as well as the feasibility of adapting inverse dynamics to Kinect-identified joint centers. The purpose of the current study is to first propose a method to align the coordinate system of the Kinect sensor with respect to the global coordinate system of a motion tracking system, and then to examine the accuracy of the Kinect sensor-identified coordinates of joint locations during 8 standing and 8 sitting postures of daily activities. The results indicate the proposed alignment method can effectively align the Kinect sensor with respect to the motion tracking system. The accuracy level of the Kinect-identified joint center location is posture-dependent and joint-dependent. For upright standing posture, the average error across all the participants and all Kinect-identified joint centers is 76 mm and 87 mm for the first and second generation Kinect sensor, respectively. In general, standing postures can be identified with better accuracy than sitting postures, and the identification accuracy of the joints of the upper extremities is better than for the lower extremities. This result may provide some information regarding the feasibility of using the Kinect sensor in future studies.}, journal={APPLIED ERGONOMICS}, author={Xu, Xu and McGorry, Raymond W.}, year={2015}, month={Jul}, pages={47–54} }
 @article{banks_chang_xu_chang_2015, title={Using horizontal heel displacement to identify heel strike instants in normal gait}, volume={42}, ISSN={["1879-2219"]}, DOI={10.1016/j.gaitpost.2015.03.015}, abstractNote={Heel strike instants are an important component of gait analyses, yet accurate detection can be difficult without a force plate. This paper presents two novel techniques for kinematic heel strike instant (kHSI) detection which examined maximal resultant horizontal heel displacement (HHD). Each of these HHD techniques calculates HHD from a selected reference location of either the stance ankle or stance heel to the swing heel. The proposed techniques, along with other previously established techniques, were validated against a 10 N force plate threshold. Fifty-four healthy adults walked overground at both normal and fast speeds while wearing athletic shoes. The reported true and absolute errors were as low as 3.2 (4.4) and 5.7 (3.4) ms, respectively, across 8678 kHSI when using the stance ankle as a reference, which significantly outperformed (p < 0.0001) the established techniques. Gait speed was shown to have a significant effect (p < 0.0001) on HHD-determined kHSI, as well as the three other techniques evaluated, highlighting the need for condition-specific identification of kHSI.}, number={1}, journal={GAIT & POSTURE}, author={Banks, Jacob J. and Chang, Wen-Ruey and Xu, Xu and Chang, Chien-Chi}, year={2015}, month={Jun}, pages={101–103} }
 @article{xu_mcgorry_lin_2014, title={A regression model predicting isometric shoulder muscle activities from arm postures and shoulder joint moments}, volume={24}, ISSN={["1873-5711"]}, DOI={10.1016/j.jelekin.2014.02.004}, abstractNote={Tissue overloading is a major contributor to shoulder musculoskeletal injuries. Previous studies attempted to use regression-based methods to predict muscle activities from shoulder kinematics and shoulder kinetics. While a regression-based method can address co-contraction of the antagonist muscles as opposed to the optimization method, most of these regression models were based on limited shoulder postures. The purpose of this study was to develop a set of regression equations to predict the 10th percentile, the median, and the 90th percentile of normalized electromyography (nEMG) activities from shoulder postures and net shoulder moments. Forty participants generated various 3-D shoulder moments at 96 static postures. The nEMG of 16 shoulder muscles was measured and the 3-D net shoulder moment was calculated using a static biomechanical model. A stepwise regression was used to derive the regression equations. The results indicated the measured range of the 3-D shoulder moment in this study was similar to those observed during work requiring light physical capacity. The r2 of all the regression equations ranged between 0.228 and 0.818. For the median of the nEMG, the average r2 among all 16 muscles was 0.645, and the five muscles with the greatest r2 were the three deltoids, supraspinatus, and infraspinatus. The results can be used by practitioners to estimate the range of the shoulder muscle activities given a specific arm posture and net shoulder moment.}, number={3}, journal={JOURNAL OF ELECTROMYOGRAPHY AND KINESIOLOGY}, author={Xu, Xu and McGorry, Raymond W. and Lin, Jia-Hua}, year={2014}, month={Jun}, pages={419–429} }
 @article{xu_lin_mcgorry_2014, title={A regression-based 3-D shoulder rhythm}, volume={47}, ISSN={["1873-2380"]}, DOI={10.1016/j.jbiomech.2014.01.043}, abstractNote={In biomechanical modeling of the shoulder, it is important to know the orientation of each bone in the shoulder girdle when estimating the loads on each musculoskeletal element. However, because of the soft tissue overlying the bones, it is difficult to accurately derive the orientation of the clavicle and scapula using surface markers during dynamic movement. The purpose of this study is to develop two regression models which predict the orientation of the clavicle and the scapula. The first regression model uses humerus orientation and individual factors such as age, gender, and anthropometry data as the predictors. The second regression model includes only the humerus orientation as the predictor. Thirty-eight participants performed 118 static postures covering the volume of the right hand reach. The orientation of the thorax, clavicle, scapula and humerus were measured with a motion tracking system. Regression analysis was performed on the Euler angles decomposed from the orientation of each bone from 26 randomly selected participants. The regression models were then validated with the remaining 12 participants. The results indicate that for the first model, the r(2) of the predicted orientation of the clavicle and the scapula ranged between 0.31 and 0.65, and the RMSE obtained from the validation dataset ranged from 6.92° to 10.39°. For the second model, the r(2) ranged between 0.19 and 0.57, and the RMSE obtained from the validation dataset ranged from 6.62° and 11.13°. The derived regression-based shoulder rhythm could be useful in future biomechanical modeling of the shoulder.}, number={5}, journal={JOURNAL OF BIOMECHANICS}, author={Xu, Xu and Lin, Jia-hua and McGorry, Raymond W.}, year={2014}, month={Mar}, pages={1206–1210} }
 @article{qin_trudeau_buchholz_katz_xu_dennerlein_2014, title={Joint Contribution to Fingertip Movement During a Number Entry Task: An Application of Jacobian Matrix}, volume={30}, ISSN={["1543-2688"]}, DOI={10.1123/jab.2013-0093}, abstractNote={Upper extremity kinematics during keyboard use is associated with musculoskeletal health among computer users; however, specific kinematics patterns are unclear. This study aimed to determine the dynamic roles of the shoulder, elbow, wrist and metacarpophalangeal (MCP) joints during a number entry task. Six subjects typed in phone numbers using their right index finger on a stand-alone numeric keypad. The contribution of each joint of the upper extremity to the fingertip movement during the task was calculated from the joint angle trajectory and the Jacobian matrix of a nine-degree-of-freedom kinematic representation of the finger, hand, forearm and upper arm. The results indicated that in the vertical direction where the greatest fingertip movement occurred, the MCP, wrist, elbow (including forearm) and shoulder joint contributed 10.2%, 55.6%, 27.7% and 6.5%, respectively, to the downward motion of the index finger averaged across subjects. The results demonstrated that the wrist and elbow contribute the most to the fingertip vertical movement, indicating that they play a major role in the keying motion and have a dynamic load beyond maintaining posture.}, number={2}, journal={JOURNAL OF APPLIED BIOMECHANICS}, author={Qin, Jin and Trudeau, Matthieu and Buchholz, Bryan and Katz, Jeffrey N. and Xu, Xu and Dennerlein, Jack T.}, year={2014}, month={Apr}, pages={338–342} }
 @article{qin_lin_buchholz_xu_2014, title={Shoulder muscle fatigue development in young and older female adults during a repetitive manual task}, volume={57}, ISSN={["1366-5847"]}, DOI={10.1080/00140139.2014.914576}, abstractNote={Age may modify the association between occupational physical demand and muscle loading, and ultimately increase the risk of musculoskeletal disorders. The goal of this study was to investigate age-related differences in shoulder muscle fatigue development during a repetitive manual task. Twenty participants in two age groups completed an 80-minute simulated low-intensity assembly task. Electromyographic (EMG) manifestation of muscle fatigue was observed in the upper trapezius, deltoid and infraspinatus muscles in both age groups, and coincided with an increase in the subjective ratings of perceived exertions. Compared with the younger group, older group showed a more monotonic decrease in EMG power frequency in the upper trapezius and deltoid muscles. However, the age-related difference in EMG amplitude was less consistent. Relative rest time of the upper trapezius muscle in the older group was less than the young group throughout the task. The observed patterns of EMG measures suggest that older participants may have disadvantages in fatigue resistance in the upper trapezius and posterior deltoid muscles during the simulated repetitive manual task. Practitioner Summary: A rapidly ageing workforce in the USA and other countries poses new challenges for preventing work-related injuries. This study showed that during an 80-minute repetitive light manual work, older adults exhibited more consistent patterns of electromyographic manifestation of shoulder muscle fatigue and less rest in the upper trapezius muscle than young adults.}, number={8}, journal={ERGONOMICS}, author={Qin, Jin and Lin, Jia-Hua and Buchholz, Bryan and Xu, Xu}, year={2014}, month={Aug}, pages={1201–1212} }
 @article{xu_mcgorry_lin_2014, title={The accuracy of an external frame using ISB recommended rotation sequence to define shoulder joint angle}, volume={39}, ISSN={["1879-2219"]}, DOI={10.1016/j.gaitpost.2013.08.032}, abstractNote={When investigating shoulder kinematics, it may be necessary to limit shoulder joint angles at a specific level. Previous studies used external frames or external surfaces to assist the participant to reach the shoulder joint angles of interest. The accuracy of these methods, however, has not yet been investigated. In the current study, an external frame was designed to assist in maintaining specific shoulder postures in a wide range. The three degrees of freedom of rotation of the proposed frame were designed to be consistent with the description of shoulder joint angles recommended by the International Society of Biomechanics. Six participants used this frame to perform 118 different shoulder postures. The reference joint angles measured by a motion tracking system were compared with the frame-defined angles. The angle differences among all the participants ranged from 12.7° to 85.6°, with an average of 32.2° (SD 15.1°) across all postures. For the postures with elevation angles on or below horizontal, the average angle difference was 23.7° (SD 8.5°). Findings suggest that errors exist when using an external frame to assist in reaching specific shoulder postures. Error is minimized at elevation angles close to −30°, and the performance is poor for extreme shoulder postures.}, number={1}, journal={GAIT & POSTURE}, author={Xu, Xu and McGorry, Raymond W. and Lin, Jia-hua}, year={2014}, month={Jan}, pages={662–668} }
 @article{xu_qin_zhang_lin_2014, title={The effect of age on the hand movement time during machine paced assembly tasks for female workers}, volume={44}, ISSN={["1872-8219"]}, DOI={10.1016/j.ergon.2013.11.010}, abstractNote={The share of older adults in the workforce is increasing in many countries. In the manufacturing industry a high proportion of assembly tasks are machine paced. Previous studies have shown that older adults tend to have longer movement times than younger adults when working at a self-selected pace. However, it is unclear whether older adults can obtain the same hand movement time as a younger group when performing machine paced work at the assembly line. In the current study, 10 older and 10 younger female participants performed simulated light-duty assembly tasks during which the hand movement times were recorded. The results showed that the older participants were capable of working at the set pace and there was no significant difference between age groups in hand movement times (989.9 msec vs. 986.6 msec, p = 0.5647). A likely explanation to the results is that the older participant had to work closer to their physical limits or capacity in order to compensate for the age effect on movement time. This study provided some preliminary quantitative data describing the hand movement time for younger and older female adults during machine paced assembly work. The results showed that age did not have a significant effect on hand movement time. Such results may help in adapting workplaces and work tasks to accommodate the needs of an aging workforce.}, number={1}, journal={INTERNATIONAL JOURNAL OF INDUSTRIAL ERGONOMICS}, author={Xu, Xu and Qin, Jin and Zhang, Tao and Lin, Jia-Hua}, year={2014}, month={Jan}, pages={148–152} }
 @article{qin_lin_faber_buchholz_xu_2014, title={Upper extremity kinematic and kinetic adaptations during a fatiguing repetitive task}, volume={24}, ISSN={["1873-5711"]}, DOI={10.1016/j.jelekin.2014.02.001}, abstractNote={Repetitive low-force contractions are common in the workplace and yet can lead to muscle fatigue and work-related musculoskeletal disorders. The current study aimed to investigate potential motion adaptations during a simulated repetitive light assembly work task designed to fatigue the shoulder region, focusing on changes over time and age-related group differences. Ten younger and ten older participants performed four 20-min task sessions separated by short breaks. Mean and variability of joint angles and scapular elevation, joint net moments for the shoulder, elbow, and wrist were calculated from upper extremity kinematics recorded by a motion tracking system. Results showed that joint angle and joint torque decreased across sessions and across multiple joints and segments. Increased kinematic variability over time was observed in the shoulder joint; however, decreased kinematic variability over time was seen in the more distal part of the upper limb. The changes of motion adaptations were sensitive to the task-break schedule. The results suggested that kinematic and kinetic adaptations occurred to reduce the biomechanical loading on the fatigued shoulder region. In addition, the kinematic and kinetic responses at the elbow and wrist joints also changed, possibly to compensate for the increased variability caused by the shoulder joint while still maintaining task requirements. These motion strategies in responses to muscle fatigue were similar between two age groups although the older group showed more effort in adaptation than the younger in terms of magnitude and affected body parts.}, number={3}, journal={JOURNAL OF ELECTROMYOGRAPHY AND KINESIOLOGY}, author={Qin, Jin and Lin, Jia-Hua and Faber, Gert S. and Buchholz, Bryan and Xu, Xu}, year={2014}, month={Jun}, pages={404–411} }
 @article{xu_qin_catena_faber_lin_2013, title={Effect of aging on inter-joint synergies during machine-paced assembly tasks}, volume={231}, ISSN={["1432-1106"]}, DOI={10.1007/s00221-013-3688-9}, abstractNote={In recent years, uncontrolled manifold (UCM) analysis has emerged as an important method to study variability of human movements. The current study investigated the upper extremity movements during typical assembly tasks using the framework of the UCM analysis. Younger and older participants performed machine-paced assembly tasks, while the kinematics of upper extremities were recorded using a motion tracking system. The upper extremity was modeled as a 7 degrees-of-freedom system. The variance of joint angles within the UCM space (V UCM) and the variance perpendicular to the UCM space (V ORT) were analyzed. The results indicated that V UCM were not significantly different for the older and younger groups. For the older group, V ORT was significantly less than the younger group and resulted in less total variance (V TOT) and a better synergy level (Z ΔV ). Therefore, the synergies of upper extremity movement may not be impaired for machine-paced tasks as people age. While current results showed a different effect of aging on the synergies of body movement compared with one previous study, they were in line with a recently proposed theory that for natural tasks, aging people did not have impairment in the ability to organize upper extremity movement into synergies.}, number={2}, journal={EXPERIMENTAL BRAIN RESEARCH}, author={Xu, Xu and Qin, Jin and Catena, Robert D. and Faber, Gert S. and Lin, Jia-Hua}, year={2013}, month={Nov}, pages={249–256} }
 @article{xu_lin_boyer_2013, title={Shoulder joint loading and posture during medicine cart pushing task}, volume={10}, DOI={10.1080/15459624.2013.803417}, abstractNote={Excessive physical loads and awkward shoulder postures during pushing and pulling are risk factors for shoulder pain. Pushing a medicine cart is a major component of a work shift for nurses and medical assistants in hospitals and other health care facilities. A laboratory experiment was conducted to examine the effects of common factors (e.g., lane congestion, cart load stability, floor surface friction) on shoulder joint moment and shoulder elevation angle of participants during cart pushing. Participants pushed a medicine cart on straight tracks and turning around right-angle corners. Peak shoulder joint moments reached 25.1 Nm, 20.3 Nm, and 26.8 Nm for initial, transition, and turning phases of the pushing tasks, indicating that shoulder joint loading while pushing a medical cart is comparable to levels previously reported from heavy manual activities encountered in industry (e.g., garbage collection). Also, except for user experience, all other main study factors, including congestion level, cart load stability, location of transition strip, shoulder tendency, surface friction, and handedness, significantly influenced shoulder joint moment and shoulder elevation angle. The findings provide a better understanding of shoulder exposures associated with medicine cart operations and may be helpful in designing and optimizing the physical environment where medicine carts are used.}, number={8}, journal={Journal of Occupational and Environmental Hygiene}, author={Xu, Xu and Lin, J. H. and Boyer, J.}, year={2013}, pages={446–454} }
 @article{xu_faber_kingma_chang_hsiang_2013, title={The error of L5/S1 joint moment calculation in a body-centered non-inertial reference frame when the fictitious force is ignored}, volume={46}, ISSN={["0021-9290"]}, DOI={10.1016/j.jbiomech.2013.05.012}, abstractNote={In ergonomics studies, linked segment models are commonly used for estimating dynamic L5/S1 joint moments during lifting tasks. The kinematics data input to these models are with respect to an arbitrary stationary reference frame. However, a body-centered reference frame, which is defined using the position and the orientation of human body segments, is sometimes used to conveniently identify the location of the load relative to the body. When a body-centered reference frame is moving with the body, it is a non-inertial reference frame and fictitious force exists. Directly applying a linked segment model to the kinematics data with respect to a body-centered non-inertial reference frame will ignore the effect of this fictitious force and introduce errors during L5/S1 moment estimation. In the current study, various lifting tasks were performed in the laboratory environment. The L5/S1 joint moments during the lifting tasks were calculated by a linked segment model with respect to a stationary reference frame and to a body-centered non-inertial reference frame. The results indicate that applying a linked segment model with respect to a body-centered non-inertial reference frame will result in overestimating the peak L5/S1 joint moments of the coronal plane, sagittal plane, and transverse plane during lifting tasks by 78%, 2%, and 59% on average, respectively. The instant when the peak moment occurred was delayed by 0.13, 0.03, and 0.09 s on average, correspondingly for the three planes. The root-mean-square errors of the L5/S1 joint moment for the three planes are 21 Nm, 19 Nm, and 9 Nm, correspondingly.}, number={11}, journal={JOURNAL OF BIOMECHANICS}, author={Xu, Xu and Faber, Gert S. and Kingma, Idsart and Chang, Chien-Chi and Hsiang, Simon M.}, year={2013}, month={Jul}, pages={1943–1947} }
 @article{chang_xu_faber_kingma_dennerlein_2012, title={Assessing manual lifting tasks based on segment angle interpolations}, volume={41 Suppl 1}, journal={Work (Reading, Mass.)}, author={Chang, C. C. and Xu, X. and Faber, G. S. and Kingma, I. and Dennerlein, J. T.}, year={2012}, pages={2360–2363} }
 @article{xu_lin_mcgorry_2012, title={Coordinate transformation between shoulder kinematic descriptions in the Holzbaur et al. model and ISB sequence}, volume={45}, ISSN={["1873-2380"]}, DOI={10.1016/j.jbiomech.2012.08.018}, abstractNote={Holzbaur et al. (2005) proposed a comprehensive 3-D biomechanical upper extremity model. Since then, this model has been adopted by many other studies for kinetic and kinematic analysis of the shoulder joint. Because of the 3-D anatomical structure, three angles are necessary to define or describe shoulder kinematics. In the Holzbaur et al. model, the three angles are shoulder elevation, elevation angle, and shoulder rotation. The computational implementation of the elevation angle degree of freedom is considered in a different way than described in the recommendation of the International Society of Biomechanics (ISB). This paper presents an analysis of the transformation between the coordinates of the shoulder kinematic defined in the Holzbaur et al. upper extremity model and those defined by the ISB. The results of this study could be used for comparing the coordinates between the different descriptions of the shoulder kinematics.}, number={15}, journal={JOURNAL OF BIOMECHANICS}, author={Xu, Xu and Lin, Jia-Hua and McGorry, Raymond W.}, year={2012}, month={Oct}, pages={2715–2718} }
 @article{xu_chang_faber_kingma_dennerlein_2012, title={Estimating 3-D L5/S1 Moments During Manual Lifting Using a Video Coding System: Validity and Interrater Reliability}, volume={54}, ISSN={["0018-7208"]}, DOI={10.1177/0018720812441945}, abstractNote={Objective: The aim of the study was to investigate the validity and interrater reliability of using a proposed video coding system to estimate the dynamical 3-D L5/S1 joint moment on the basis of four key frames from video clips of asymmetric lifting tasks. }, number={6}, journal={HUMAN FACTORS}, author={Xu, Xu and Chang, Chien-Chi and Faber, Gert S. and Kingma, Idsart and Dennerlein, Jack T.}, year={2012}, month={Dec}, pages={1053–1065} }
 @article{xu_chang_faber_kingma_dennerlein_2012, title={Estimation of 3-D peak L5/S1 joint moment during asymmetric lifting tasks with cubic spline interpolation of segment Euler angles}, volume={43}, ISSN={["1872-9126"]}, DOI={10.1016/j.apergo.2011.04.002}, abstractNote={Previous research proposed a method using interpolation of the joint angles in key frames extracted from a field-survey video to estimate the dynamic L5/S1 joint loading for symmetric lifting tasks. The advantage of this method is that there is no need to use unwieldy equipment for capturing full body movement for the lifting tasks. The current research extends this method to asymmetric lifting tasks. The results indicate that 4-point cubic spline interpolation of segment Euler angles combined with a biomechanical model can provide a good estimation of 3-D peak L5/S1 joint moments for asymmetric lifting tasks. The average absolute error in the coronal, sagittal, and transverse planes with respect to the local pelvis axes was 16Nm, 22Nm, and 11Nm, respectively. It was also found that the dynamic component of the peak L5/S1 joint moment was not monotonously convergent when the number of interpolation points was increased. These results can be helpful for developing applied ergonomic field-survey tools such as video bases systems for estimating L5/S1 moments of manual materials handling tasks.}, number={1}, journal={APPLIED ERGONOMICS}, author={Xu, Xu and Chang, Chien-Chi and Faber, Gert S. and Kingma, Idsart and Dennerlein, Jack T.}, year={2012}, month={Jan}, pages={115–120} }
 @article{xu_lin_li_tan_2012, title={Transformation between different local coordinate systems of the scapula}, volume={45}, ISSN={["0021-9290"]}, DOI={10.1016/j.jbiomech.2012.08.021}, abstractNote={The existence of multiple local coordinate systems (LCSs) for the scapula makes it difficult to compare the kinematics of the scapula across various studies and reports. This study aimed to build transformation matrices between different LCSs for the scapula and to provide the coordinates of previously measured muscles and ligaments around the scapula with respect to the International Society of Biomechanics (ISB) recommended LCS. The bony landmarks necessary for building various local coordinate systems were digitized on 13 CT scanned scapulae. The LCSs were built based on the digitized bony landmarks and then used for calculating the transformation equations. The approximate coordinates of 28 muscles and ligaments of the scapula were expressed with respect to the ISB-recommended LCS using the derived transformation equations. The results of this study may be used for the comparison of scapula kinematics data with respect to various LCSs and for building a scapula biomechanical model with respect to ISB-recommended LCS.}, number={15}, journal={JOURNAL OF BIOMECHANICS}, author={Xu, Xu and Lin, Jia-Hua and Li, Kang and Tan, Virak}, year={2012}, month={Oct}, pages={2724–2727} }
 @article{xu_chang_lu_2012, title={Two linear regression models predicting cumulative dynamic L5/S1 joint moment during a range of lifting tasks based on static postures}, volume={55}, ISSN={["1366-5847"]}, DOI={10.1080/00140139.2012.693627}, abstractNote={Previous studies have indicated that cumulative L5/S1 joint load is a potential risk factor for low back pain. The assessment of cumulative L5/S1 joint load during a field study is challenging due to the difficulty of continuously monitoring the dynamic joint load. This study proposes two regression models predicting cumulative dynamic L5/S1 joint moment based on the static L5/S1 joint moment of a lifting task at lift-off and set-down and the lift duration. Twelve men performed lifting tasks at varying lifting ranges and asymmetric angles in a laboratory environment. The cumulative L5/S1 joint moment was calculated from continuous dynamic L5/S1 moments as the reference for comparison. The static L5/S1 joint moments at lift-off and set-down were measured for the two regression models. The prediction error of the cumulative L5/S1 joint moment was 21±14 Nm × s (12% of the measured cumulative L5/S1 joint moment) and 14±9 Nm × s (8%) for the first and the second models, respectively. Practitioner Summary: The proposed regression models may provide a practical approach for predicting the cumulative dynamic L5/S1 joint loading of a lifting task for field studies since it requires only the lifting duration and the static moments at the lift-off and/or set-down instants of the lift.}, number={9}, journal={ERGONOMICS}, author={Xu, Xu and Chang, Chien-Chi and Lu, Ming-Lun}, year={2012}, pages={1093–1103} }
 @article{coenen_kingma_boot_faber_xu_bongers_dieen_2011, title={Estimation of low back moments from video analysis: A validation study}, volume={44}, ISSN={["0021-9290"]}, DOI={10.1016/j.jbiomech.2011.07.005}, abstractNote={This study aimed to develop, compare and validate two versions of a video analysis method for assessment of low back moments during occupational lifting tasks since for epidemiological studies and ergonomic practice relatively cheap and easily applicable methods to assess low back loads are needed. Ten healthy subjects participated in a protocol comprising 12 lifting conditions. Low back moments were assessed using two variants of a video analysis method and a lab-based reference method. Repeated measures ANOVAs showed no overall differences in peak moments between the two versions of the video analysis method and the reference method. However, two conditions showed a minor overestimation of one of the video analysis method moments. Standard deviations were considerable suggesting that errors in the video analysis were random. Furthermore, there was a small underestimation of dynamic components and overestimation of the static components of the moments. Intraclass correlations coefficients for peak moments showed high correspondence (>0.85) of the video analyses with the reference method. It is concluded that, when a sufficient number of measurements can be taken, the video analysis method for assessment of low back loads during lifting tasks provides valid estimates of low back moments in ergonomic practice and epidemiological studies for lifts up to a moderate level of asymmetry.}, number={13}, journal={JOURNAL OF BIOMECHANICS}, author={Coenen, Pieter and Kingma, Idsart and Boot, Cecile R. L. and Faber, Gert S. and Xu, Xu and Bongers, Paulien M. and Dieen, Jaap H.}, year={2011}, month={Sep}, pages={2369–2375} }
 @inbook{xu_chang_faber_kingma_dennerlein_2011, title={Postural Observation of Shoulder Flexion during Asymmetric Lifting Tasks}, ISBN={9783642217982 9783642217999}, ISSN={0302-9743 1611-3349}, url={http://dx.doi.org/10.1007/978-3-642-21799-9_26}, DOI={10.1007/978-3-642-21799-9_26}, abstractNote={This study was to evaluate the observation error of the shoulder flexion angle during an asymmetric lifting task. The results indicated the average absolute estimate error was 14.7 degrees and the correlation coefficient between the measured and estimated shoulder flexion was 0.91. The observation error may be due to the arm abduction.}, booktitle={Digital Human Modeling}, publisher={Springer Berlin Heidelberg}, author={Xu, Xu and Chang, Chien-Chi and Faber, Gert S. and Kingma, Idsart and Dennerlein, Jack T.}, year={2011}, pages={228–230} }
 @article{xu_chang_faber_kingma_dennerlein_2011, title={The Validity and Interrater Reliability of Video-Based Posture Observation During Asymmetric Lifting Tasks}, volume={53}, ISSN={["0018-7208"]}, DOI={10.1177/0018720811410976}, abstractNote={Objective: The objective was to evaluate the validity and interrater reliability of a video-based posture observation method for the major body segment angles during asymmetric lifting tasks.}, number={4}, journal={HUMAN FACTORS}, author={Xu, Xu and Chang, Chien-chi and Faber, Gert S. and Kingma, Idsart and Dennerlein, Jack T.}, year={2011}, month={Aug}, pages={371–382} }
 @article{xu_hsiang_mirka_2010, title={An empirical validation of a base-excitation model to predict harvestable energy from a suspended-load backpack system}, volume={11}, ISSN={["1464-536X"]}, DOI={10.1080/14639220903373839}, abstractNote={Suspended-load backpacks have been proposed as a way to provide power for small electronic devices by capturing the mechanical energy generated by the vertical movement of the suspended load and converting it into electrical energy. The aim of the current study was to build a base excitation model able to predict the relative velocity of the load (an index of the amount of harvestable energy of such a system) using as inputs the mass of the suspended load, the walking speed and the leg length of the user. Nine human participants walked on a treadmill under two load conditions (15.8 kg and 22.6 kg load) and three walking speed conditions (1.16 m/s, 1.43 m/s and 1.70 m/s). The predictions of the load velocity by the base-excitation model under these conditions were then compared with the measured load velocity. The results of this study showed a moderately strong correlation (0.76) between the root mean square of the predicted and measured relative velocity of the load, and the average absolute error of these predictions was 24.2%. These results provide support for the utility of this approach and also provide motivation for further refinement of the base excitation model for the prediction of the amount of energy able to be harvested from suspended-load backpack systems.}, number={6}, journal={THEORETICAL ISSUES IN ERGONOMICS SCIENCE}, author={Xu, X. and Hsiang, S. and Mirka, G.}, year={2010}, pages={546–560} }
 @article{xu_chang_faber_kingma_dennerlein_2010, title={Comparing polynomial and cubic spline interpolation of segment angles for estimating L5/S1 net moment during symmetric lifting tasks}, volume={43}, ISSN={["0021-9290"]}, DOI={10.1016/j.jbiomech.2009.09.044}, abstractNote={Simple video-based methods previously proposed for field research to estimate L5/S1 net moments during real-world manual materials handling rely on polynomial interpolation on the joint angles from key frames extracted from video recordings; however, polynomial interpolations may not converge as the number of interpolation points increases. Therefore, we compared L5/S1 net moments calculated from continuous kinematic measurements to those calculated from both polynomial and cubic spline interpolation on body segments angles during lifting tasks. For small number of interpolation points (<6) the error in the predicted moment from both the spline and polynomial fits decreased with the increase in the number of interpolation points; however, above 6 interpolation points error for the polynomial fits started to increase while the error from the spline fit continued to decrease. These results suggest that cubic spline interpolation on body segments angles provides a more robust basis for calculating L5/S1 net moment from a few key video frames.}, number={3}, journal={JOURNAL OF BIOMECHANICS}, author={Xu, Xu and Chang, Chien-Chi and Faber, Gert S. and Kingma, Idsart and Dennerlein, Jack T.}, year={2010}, month={Feb}, pages={583–586} }
 @article{xu_chang_faber_kingma_dennerlein_2010, title={Interpolation of segment Euler angles can provide a robust estimation of segment angular trajectories during asymmetric lifting tasks}, volume={43}, ISSN={["1873-2380"]}, DOI={10.1016/j.jbiomech.2010.03.010}, abstractNote={Video-based field methods that estimate L5/S1 net joint moments from kinematics based on interpolation in the sagittal plane of joint angles alone can introduce a significant error on the interpolated joint angular trajectory when applied to asymmetric dynamic lifts. Our goal was to evaluate interpolation of segment Euler angles for a wide range of dynamic asymmetric lifting tasks using cubic spline methods by comparing the interpolated values with the continuous measured ones. For most body segments, the estimated trajectories of segment Euler angles have less than 5° RMSE (in each dimension) with 5-point cubic spline interpolation when there is no measurement error of interpolation points. Sensitivity analysis indicates that when the measurement error exists, the root mean square error (RMSE) of estimated trajectories increases. Comparison among different lifting conditions showed that lifting a load from a high initial position yielded a smaller RMSE than lifting from a low initial position. In conclusion, interpolation of segment Euler angles can provide a robust estimation of segment angular trajectories during asymmetric lifting when measurement error of interpolation points can be controlled at a low level.}, number={10}, journal={JOURNAL OF BIOMECHANICS}, author={Xu, Xu and Chang, Chien-Chi and Faber, Gert S. and Kingma, Idsart and Dennerlein, Jack T.}, year={2010}, month={Jul}, pages={2043–2048} }
 @article{chang_mcgorry_lin_xu_hsiang_2010, title={Prediction accuracy in estimating joint angle trajectories using a video posture coding method for sagittal lifting tasks}, volume={53}, ISSN={["0014-0139"]}, DOI={10.1080/00140139.2010.489963}, abstractNote={This study investigated prediction accuracy of a video posture coding method for lifting joint trajectory estimation. From three filming angles, the coder selected four key snapshots, identified joint angles and then a prediction program estimated the joint trajectories over the course of a lift. Results revealed a limited range of differences of joint angles (elbow, shoulder, hip, knee, ankle) between the manual coding method and the electromagnetic motion tracking system approach. Lifting range significantly affected estimate accuracy for all joints and camcorder filming angle had a significant effect on all joints but the hip. Joint trajectory predictions were more accurate for knuckle-to-shoulder lifts than for floor-to-shoulder or floor-to-knuckle lifts with average root mean square errors (RMSE) of 8.65°, 11.15° and 11.93°, respectively. Accuracy was also greater for the filming angles orthogonal to the participant's sagittal plane (RMSE = 9.97°) as compared to filming angles of 45° (RMSE = 11.01°) or 135° (10.71°). The effects of lifting speed and loading conditions were minimal. To further increase prediction accuracy, improved prediction algorithms and/or better posture matching methods should be investigated. Statement of Relevance: Observation and classification of postures are common steps in risk assessment of manual materials handling tasks. The ability to accurately predict lifting patterns through video coding can provide ergonomists with greater resolution in characterising or assessing the lifting tasks than evaluation based solely on sampling with a single lifting posture event.}, number={8}, journal={ERGONOMICS}, author={Chang, Chien-Chi and McGorry, Raymond W. and Lin, Jia-hua and Xu, Xu and Hsiang, Simon M.}, year={2010}, pages={1039–1047} }
 @article{xu_hsiang_mirka_2009, title={The effects of a suspended-load backpack on gait}, volume={29}, ISSN={["1879-2219"]}, DOI={10.1016/j.gaitpost.2008.06.008}, abstractNote={A suspended-load backpack is a device that is designed to capture the mechanical energy created as a suspended backpack load oscillates vertically on the back during gait. The objective of the current study was to evaluate the effect of a suspended-load backpack system on selected temporal and kinetics parameters describing gait. Nine male participants carried a suspended-load backpack as they walked on an instrumented treadmill with varied levels of load (no backpack, 22.5 kg, and 29.3 kg) and walking speed (1.16 m/s, 1.43 m/s, 1.70 m/s). As the participants performed this treadmill task, ground reaction forces were collected from an instrumented treadmill system. From these data, temporal variables (cycle time, single support time, and double support time) and kinetic variables (normalized weight acceptance force, normalized push-off force, and normalized mid-stance force) were derived. The results showed that the response of the temporal variables were consistent with previous studies of conventional (i.e. stable load) backpacks. The response of the normalized push-off force, however, showed that increasing walking speed significantly (p<0.05) decreased the magnitude of this force, a result contrary to the literature concerning conventional backpacks where this force has been shown to significantly increase. Further evaluation revealed that this reduction in force was the result of a phase shift between the movement of the carried load and the movement of the torso. This suggests that the motion of the load in a suspended-load backpack influences the gait biomechanics and should be considered as this technology advances.}, number={1}, journal={GAIT & POSTURE}, author={Xu, Xu and Hsiang, Simon M. and Mirka, Gary A.}, year={2009}, month={Jan}, pages={151–153} }
 @article{xu_hsiang_mirka_2008, title={Coordination indices between lifting kinematics and kinetics}, volume={38}, ISSN={["1872-8219"]}, DOI={10.1016/j.ergon.2008.02.008}, abstractNote={During a lifting task the movement of the trunk can account for the majority of the external moment about the ankle. Though the angle of trunk flexion and the external moment about the ankles are roughly correlated, this correlation can be reduced by various segmental dynamics and momentums with the upper/lower extremities. Two methods are proposed in this technical note for describing the relationship between the kinematics and the kinetics of a lifting motion. The first relies on the phase plane analysis technique and explores the relative phase angle between the kinematic characteristics of lifting motion (i.e., trunk motion in the sagittal plane) and the kinetic characteristics of lifting motion (i.e., the net external moment). The second technique employs the moving correlation technique that assesses the level of coordination between the net external moment and the angle of the torso in the sagittal plane. In this paper, these methods are applied to a dataset of lifting motions of obese and normal weight participants to explore the utility of these modeling approaches on the assessment of potential risk in the lifting task due to obesity. Quantifying the coordination between the movement of the trunk and the net external moment can be helpful in understanding the lifting techniques that may place the lifter at higher risk of developing a low back injury.}, number={11-12}, journal={INTERNATIONAL JOURNAL OF INDUSTRIAL ERGONOMICS}, author={Xu, Xu and Hsiang, Simon M. and Mirka, Gary A.}, year={2008}, pages={1062–1066} }
 @article{xu_mirka_hsiang_2008, title={The effects of obesity on lifting performance}, volume={39}, ISSN={["0003-6870"]}, DOI={10.1016/j.apergo.2007.02.001}, abstractNote={Obesity in the workforce is a growing problem worldwide. While the implications of this trend for biomechanical loading of the musculoskeletal system seem fairly straightforward, the evidence of a clear link between low back pain (LBP) and body mass index (BMI) (calculated as whole body mass in kilograms divided by the square of stature in meters) has not been shown in the epidemiology literature addressing this topic. The approach pursued in the current study was to evaluate the lifting kinematics and ground reaction forces of a group of 12 subjects -- six with a BMI of less than 25 kg/m(2) (normal weight) and six with a BMI of greater than 30 kg/m(2) (obese). These subjects performed a series of free dynamic lifting tasks with varied levels of load (10% and 25% of capacity) and symmetry (sagittally symmetric and 45 degrees asymmetric). The results showed that BMI had a significant effect (p<0.05) on trunk kinematics with the high BMI group exhibiting higher peak transverse plane (twisting) velocity (59% higher) and acceleration (57% higher), and exhibiting higher peak sagittal plane velocity (30% higher) and acceleration (51% higher). When normalized to body weight, there were no significant differences in the ground reaction forces between the two groups. This study provides quantitative data describing lifting task performance differences between people of differing BMI levels and may help to explain why there is no conclusive epidemiological evidence of a relationship between BMI and LBP.}, number={1}, journal={APPLIED ERGONOMICS}, author={Xu, Xu and Mirka, Gary A. and Hsiang, Simon M.}, year={2008}, month={Jan}, pages={93–98} }
 @article{shu_jiang_xu_mirka_2007, title={The effect of a knee support on the biomechanical response of the low back}, volume={23}, ISSN={["1065-8483"]}, DOI={10.1123/jab.23.4.275}, abstractNote={Stooping and squatting postures are seen in a number of industries (e.g., agriculture, construction) where workers must work near ground level for extended periods of time. The focus of the current research was to evaluate a knee support device designed to reduce the biomechanical loading of these postures. Ten participants performed a series of sudden loading tasks while in a semisquat posture under two conditions of knee support (no support and fully supported) and two conditions of torso flexion (45 and 60°). A weight was released into the hands of the participants who then came to steady state while maintaining the designated posture. As they performed this task, the EMG responses of the trunk extensors (multifidus and erector spinae) were collected, both during the “sudden loading” phase of the trial as well as the steady weight-holding phase of the trial. As expected, the effects of torso flexion angle showed significant decreases in the activation of the multifidus muscles with greater torso angle (indicating the initiation of the flexion–relaxation response). Interestingly, the results showed that the knee support device had no effect on the activation levels of the sampled muscles, indicating that the loss of the degree of freedom from the ankle joint during the knee support condition had no impact on trunk extensor muscle response. The a priori concern with regard to these supports was that they would tend to focus loading on the low back and therefore would not serve as a potential ergonomic solution for these stooping/semisquatting tasks. Because the results of this study did not support this concern, further development of such an intervention is underway.}, number={4}, journal={JOURNAL OF APPLIED BIOMECHANICS}, author={Shu, Yu and Jiang, Zongliang and Xu, Xu and Mirka, Gary A.}, year={2007}, month={Nov}, pages={275–281} }