Math Modeling Seminar: Full-Waveform Inversion for Medical Ultrasound Imaging

Math Modeling Seminar
Full-Waveform Inversion for Medical Ultrasound Imaging

Zoom Link here

Dr. Rehman Ali
Postdoctoral Fellow, Department of Imaging Sciences
University of Rochester

Abstract:
Ultrasound imaging is widely used in clinical practice due to its safety, accessibility, and low cost, yet it remains largely qualitative, relying on images of backscattered echoes rather than quantitative estimates of tissue properties. In this talk, I present a framework for quantitative ultrasound imaging that inverts the physics of wave propagation to reconstruct spatial maps of tissue mechanical properties from measured waveforms. I discuss how this perspective naturally leads to full-waveform inversion as a PDE-constrained optimization problem. I also present experimental results in transmission ultrasound that demonstrate MRI-like image quality in breast imaging, highlighting the potential for quantitative ultrasound to complement or reduce reliance on more expensive imaging modalities. Finally, I outline the extension of this framework to pulse-echo ultrasound, where limited-view and highly ill-posed imaging conditions motivate related wave-based inverse methods. I conclude by discussing emerging directions in physics-based modeling and inversion strategies for improving robustness and clinical applicability.

Speaker Bio:
Rehman Ali received the B.S. degree in Biomedical Engineering (2016) from the Georgia Institute of Technology (Atlanta, GA, USA) in 2016, M.S. in Computational & Mathematical Engineering (2020) and Ph.D. in Electrical Engineering (2022) at Stanford University (Palo Alto, CA, USA). His research interests include signal processing, inverse problems, and computational imaging techniques applied to medical ultrasound. He is currently an NIH K99/R00 Pathway to Independence awardee at the University of Rochester in the laboratory of Dr. Nebojsa Duric to develop waveform inversion reconstruction techniques for imaging acoustic tissue properties using ultrasound computed tomography.

Intended Audience:
Beginners, undergraduates, graduates. Those with interest in the topic.