A Machine Learning Approach to Improve Photoacoustic-Guided Surgery
Photoacoustic imaging is an emerging technique that uses light and sound to visualize anatomical structures or metal implants in the human body. When being developed to guide surgeries, one primary requirement is visualization of point-like targets, such as the circular, cross-sectional views of cylindrical needles, catheters, and brachytherapy seeds. A major challenge when visualizing these point-like targets is the presence of highly echogenic structures, causing reflection artifacts so severe that they can be mistaken for true signals. This talk describes our exploration of machine learning to identify these noise artifacts for removal. Our results demonstrate strong promise to complete this task without requiring traditional signal processing.
Muyinatu A. Lediju Bell is an Assistant Professor of Electrical & Computer Engineering with a joint appointment in the Department of Biomedical Engineering at Johns Hopkins University. She directs the Photoacoustic and Ultrasonic Systems Engineering (PULSE) Lab, a highly interdisciplinary research lab that integrates optics, acoustics, robotics, signal processing, and medical-device design to engineer and deploy innovative biomedical imaging systems that simultaneously address unmet clinical needs and significantly improve the standard of patient care. Bell completed a postdoctoral fellowship with the Engineering Research Center for Computer-Integrated Surgical Systems and Technology at Johns Hopkins University, obtained a PhD in biomedical engineering from Duke University, earned a BS in mechanical engineering (biomedical engineering minor) from the Massachusetts Institute of Technology, and spent a year abroad as a Whitaker International Fellow at the Institute of Cancer Research and Royal Marsden Hospital in the United Kingdom. She was recently honored by MIT Technology Review as one of 35 Innovators Under 35 in 2016.