@article{wu_sanders_dundar_oralkan_2023, title={Deep-Learning-Based High-Intensity Focused Ultrasound Lesion Segmentation in Multi-Wavelength Photoacoustic Imaging}, volume={10}, ISSN={["2306-5354"]}, url={https://doi.org/10.3390/bioengineering10091060}, DOI={10.3390/bioengineering10091060}, abstractNote={Photoacoustic (PA) imaging can be used to monitor high-intensity focused ultrasound (HIFU) therapies because ablation changes the optical absorption spectrum of the tissue, and this change can be detected with PA imaging. Multi-wavelength photoacoustic (MWPA) imaging makes this change easier to detect by repeating PA imaging at multiple optical wavelengths and sampling the optical absorption spectrum more thoroughly. Real-time pixel-wise classification in MWPA imaging can assist clinicians in monitoring HIFU lesion formation and will be a crucial milestone towards full HIFU therapy automation based on artificial intelligence. In this paper, we present a deep-learning-based approach to segment HIFU lesions in MWPA images. Ex vivo bovine tissue is ablated with HIFU and imaged via MWPA imaging. The acquired MWPA images are then used to train and test a convolutional neural network (CNN) for lesion segmentation. Traditional machine learning algorithms are also trained and tested to compare with the CNN, and the results show that the performance of the CNN significantly exceeds traditional machine learning algorithms. Feature selection is conducted to reduce the number of wavelengths to facilitate real-time implementation while retaining good segmentation performance. This study demonstrates the feasibility and high performance of the deep-learning-based lesion segmentation method in MWPA imaging to monitor HIFU lesion formation and the potential to implement this method in real time.}, number={9}, journal={BIOENGINEERING-BASEL}, author={Wu, Xun and Sanders, Jean L. and Dundar, M. Murat and Oralkan, Omer}, year={2023}, month={Sep} }
 @article{mahmud_seok_wu_sennik_biliroglu_adelegan_kim_jur_yamaner_oralkan_2021, title={A Low-Power Wearable E-Nose System Based on a Capacitive Micromachined Ultrasonic Transducer (CMUT) Array for Indoor VOC Monitoring}, volume={21}, url={https://doi.org/10.1109/JSEN.2021.3094125}, DOI={10.1109/JSEN.2021.3094125}, abstractNote={Volatile organic compounds (VOCs) are pervasive in the environment and their real-time continuous monitoring can facilitate better understanding of their effects on human health by combining environmental factors with physiological conditions. The scope of wearable sensors for detection of VOCs is evident as the accuracy of the sensor prediction depends on its proximity to the VOC source along with the sensitivity and selectivity of the sensor itself. In this paper, we present a low-power wearable e-nose system based on a capacitive micromachined ultrasonic transducer (CMUT) array. CMUTs offer inherent benefits of excellent mass resolution, easy array fabrication, and integration with electronics, which make them an appropriate choice as a transducer element for gravimetric e-nose systems. A 5-channel CMUT sensor array was chemically functionalized and used for the detection of four volatiles, ethanol, toluene, p-xylene, and styrene. All the channels of the sensor array achieved a resolution below 10 ppm within 0.2–3% of OSHA-PEL time-weighted average (TWA) for each volatile. For each test cycle, the maximum frequency shift, the rate of adsorption, and the rate of desorption were extracted as features. Linear discriminant analysis (LDA) was applied to visualize the discrimination performance of the sensor array. The system performance was characterized using an automated testing system. The presented sensor system can be used for identification of volatiles with suitable pattern-recognition techniques.}, number={18}, journal={IEEE Sensors Journal}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Mahmud, Marzana Mantasha and Seok, Chunkyun and Wu, Xun and Sennik, Erdem and Biliroglu, Ali Onder and Adelegan, Oluwafemi Joel and Kim, Inhwan and Jur, Jesse S. and Yamaner, Feysel Yalcin and Oralkan, Omer}, year={2021}, month={Sep}, pages={19684–19696} }
 @article{adelegan_coutant_wu_yamaner_oralkan_2021, title={Design and Fabrication of Wideband Air-Coupled Capacitive Micromachined Ultrasonic Transducers With Varying Width Annular-Ring and Spiral Cell Structures}, volume={68}, ISSN={["1525-8955"]}, url={https://doi.org/10.1109/TUFFC.2021.3076143}, DOI={10.1109/TUFFC.2021.3076143}, abstractNote={Air-coupled transducers with broad bandwidth are desired for many airborne applications, such as obstacle detection, haptic feedback, and flow metering. In this article, we present a design strategy and demonstrate a fabrication process for developing improved concentric annular- and novel spiral-shaped capacitive micromachined ultrasonic transducers (CMUTs) that can generate high output pressure and provide wide bandwidth in air. We explore the ability to implement complex geometries by photolithographic definition to improve the bandwidth of air-coupled CMUTs. The ring widths in the annular design were varied so that the device can be improved in terms of bandwidth when these rings resonate in parallel. Using the same ring width parameters for the spiral-shaped design but with a smoother transition between the ring widths along the spiral, the bandwidth of the spiral-shaped device is improved. With the reduced process complexity associated with the anodic-bonding-based fabrication process, a 25- $\mu \text{m}$ vibrating silicon plate was bonded to a borosilicate glass wafer with up to 15- $\mu \text{m}$ deep cavities. The fabricated devices show an atmospheric deflection profile that is in agreement with the FEM results to verify the vacuum sealing of the devices. The devices show a 3-dB fractional bandwidth (FBW) of 12% and 15% for spiral- and annular-shaped CMUTs, respectively. We measured a 127-dB sound pressure level at the surface of the transducers. The angular response of the fabricated CMUTs was also characterized. The results demonstrated in this article show the possibility of improving the bandwidth of air-coupled devices by exploring the flexibility in the design process associated with CMUT technology.}, number={8}, journal={IEEE TRANSACTIONS ON ULTRASONICS FERROELECTRICS AND FREQUENCY CONTROL}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Adelegan, Oluwafemi Joel and Coutant, Zachary A. and Wu, Xun and Yamaner, Feysel Yalcin and Oralkan, Omer}, year={2021}, month={Aug}, pages={2749–2759} }
 @article{sanders_biliroglu_wu_adelegan_yamaner_oralkan_2021, title={A Row-Column (RC) Addressed 2-D Capacitive Micromachined Ultrasonic Transducer (CMUT) Array on a Glass Substrate}, volume={68}, ISSN={["1525-8955"]}, url={https://doi.org/10.1109/TUFFC.2020.3014780}, DOI={10.1109/TUFFC.2020.3014780}, abstractNote={This article presents a row-column (RC) capacitive micromachined ultrasonic transducer (CMUT) array fabricated using anodic bonding on a borosilicate glass substrate. This is shown to reduce the bottom electrode-to-substrate capacitive coupling. This subsequently improves the relative response of the elements when top or bottom electrodes are used as the “signal” (active) electrode. This results in a more uniform performance for the two cases. Measured capacitance and resonant frequency, pulse-echo signal amplitude, and frequency response are presented to support this. Biasing configurations with varying ac and dc arrangements are applied and subsequently explored. Setting the net dc bias voltage across an off element to zero is found to be most effective to minimize spurious transmission. To achieve this, a custom switching circuit was designed and implemented. This circuit was also used to obtain orthogonal B-mode cross-sectional images of a rotationally asymmetric target.}, number={3}, journal={IEEE TRANSACTIONS ON ULTRASONICS FERROELECTRICS AND FREQUENCY CONTROL}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Sanders, Jean L. and Biliroglu, Ali Onder and Wu, Xun and Adelegan, Oluwafemi J. and Yamaner, Feysel Yalcin and Oralkan, Omer}, year={2021}, month={Mar}, pages={767–776} }
 @article{wu_sanders_zhang_yamaner_oralkan_2019, title={An FPGA-Based Backend System for Intravascular Photoacoustic and Ultrasound Imaging}, volume={66}, ISSN={["1525-8955"]}, url={https://doi.org/10.1109/TUFFC.2018.2881409}, DOI={10.1109/TUFFC.2018.2881409}, abstractNote={The integration of intravascular ultrasound (IVUS) and intravascular photoacoustic (IVPA) imaging produces an imaging modality with high sensitivity and specificity which is particularly needed in interventional cardiology. Conventional side-looking IVUS imaging with a single-element ultrasound (US) transducer lacks forward-viewing capability, which limits the application of this imaging mode in intravascular intervention guidance, Doppler-based flow measurement, and visualization of nearly, or totally blocked arteries. For both side-looking and forward-looking imaging, the necessity to mechanically scan the US transducer limits the imaging frame rate, and therefore, array-based solutions are desired. In this paper, we present a low-cost, compact, high-speed, and programmable imaging system based on a field-programmable gate array suitable for dual-mode forward-looking IVUS/IVPA imaging. The system has 16 US transmit and receive channels and functions in multiple modes including interleaved photoacoustic (PA) and US imaging, hardware-based high-frame-rate US imaging, software-driven US imaging, and velocity measurement. The system is implemented in the register-transfer level, and the central system controller is implemented as a finite-state machine. The system was tested with a capacitive micromachined ultrasonic transducer array. A 170-frames-per-second (FPS) US imaging frame rate is achieved in the hardware-based high-frame-rate US imaging mode while the interleaved PA and US imaging mode operates at a 60-FPS US and a laser-limited 20-FPS PA imaging frame rate. The performance of the system benefits from the flexibility and efficiency provided by the low-level implementation. The resulting system provides a convenient backend platform for research and clinical IVPA and IVUS imaging.}, number={1}, journal={IEEE TRANSACTIONS ON ULTRASONICS FERROELECTRICS AND FREQUENCY CONTROL}, publisher={Institute of Electrical and Electronics Engineers (IEEE)}, author={Wu, Xun and Sanders, Jean L. and Zhang, Xiao and Yamaner, Feysel Yalcin and Oralkan, Omer}, year={2019}, month={Jan}, pages={45–56} }
 @inproceedings{wu_sanders_dundar_oralkan_2017, title={Multi-wavelength photoacoustic imaging for monitoring lesion formation during high-intensity focused ultrasound therapy}, volume={10064}, DOI={10.1117/12.2248739}, abstractNote={Photoacoustic imaging (PAI) can be used to monitor lesion formation during high-intensity focused ultrasound (HIFU) therapy because HIFU changes the optical absorption spectrum (OAS) of the tissue. However, in traditional PAI, the change could be too subtle to be observed either because the OAS does not change very significantly at the imaging wavelength or due to low signal-to-noise ratio in general. We propose a machine-learning-based method for lesion monitoring with multi-wavelength PAI (MWPAI), where PAI is repeated at a sequence of wavelengths and a stack of multi-wavelength photoacoustic (MWPA) images is acquired. Each pixel is represented by a vector and each element in the vector reflects the optical absorption at the corresponding wavelength. Based on the MWPA images, a classifier is trained to classify pixels into two categories: ablated and non-ablated. In our experiment, we create a lesion on a block of bovine tissue with a HIFU transducer, followed by MWPAI in the 690 nm to 950 nm wavelength range, with a step size of 5 nm. In the MWPA images, some of the ablated and non-ablated pixels are cropped and fed to a neural network (NN) as training examples. The NN is then applied to several groups of MWPA images and the results show that the lesions can be identified clearly. To apply MWPAI in/near real-time, sequential feature selection is performed and the number of wavelengths is decreased from 53 to 5 while retaining adequate performance. With a fast-switching tunable laser, the method can be implemented in/near real-time.}, booktitle={Photons plus ultrasound: imaging and sensing 2017}, author={Wu, Xun and Sanders, J. and Dundar, M. and Oralkan, Omer}, year={2017} }
 @inproceedings{wu_sanders_stephens_oralkan_2016, title={Photoacoustic-imaging-based temperature monitoring for high-intensity focused ultrasound therapy}, DOI={10.1109/embc.2016.7591418}, abstractNote={Temperature monitoring during high-intensity focused ultrasound (HIFU) application is necessary to ensure effective therapy while minimizing thermal damage to adjacent tissue. In this study, we demonstrate a noninvasive approach for temperature measurement during HIFU therapy based on photoacoustic imaging (PAI). Because of the dependence of photoacoustic (PA) signal amplitude on temperature of the source tissue and the linearity of the PAI system, changes in temperature will cause changes in PA image intensity. Experiments have been conducted in ex-vivo bovine tissue to characterize the linear dependence of PA image pixel values on temperature and subsequently to convert the PA image to a real-time temperature map.}, booktitle={2016 38th annual international conference of the ieee engineering in medicine and biology society (embc)}, author={Wu, Xun and Sanders, J. L. and Stephens, D. N. and Oralkan, Omer}, year={2016}, pages={3235–3238} }
 @article{wu_kumar_oralkan_2014, title={An Ultrasound-Based Noninvasive Neural Interface to the Retina Projection Algorithm and Frontend Integrated Circuit Architecture}, ISSN={["1948-5719"]}, DOI={10.1109/ultsym.2014.0655}, abstractNote={Focused ultrasound (FUS) is emerging as a promising technology for neural stimulation. In this study, we demonstrate the algorithm and the frontend integrated circuit (IC) architecture design for an ultrasound-based noninvasive neural interface to the retina. A digital image is provided as the input to the system, and the system calculates the excitation signal for each element in a 2D transducer array. With each element being excited accordingly, the array can “project” the image onto the retina as an ultrasound field pattern (USFP). The algorithm is based on the fast Fourier transform (FFT), which makes real-time implementation feasible.}, journal={2014 IEEE INTERNATIONAL ULTRASONICS SYMPOSIUM (IUS)}, author={Wu, Xun and Kumar, Mohit and Oralkan, Omer}, year={2014}, pages={2623–2626} }