This study presents two physics-informed deep learning frameworks for improving quantitative parameter estimation in prostate and breast MRI. For 3-compartment prostate MRI, we developed PIVA, a Physics-Informed Variational Autoencoder designed to overcome the instability of conventional Non-Linear Least Squares fitting at clinically realistic signal-to-noise ratios. By embedding the known MRI signal equation within a probabilistic neural network, PIVA replaces voxel-wise curve fitting with posterior estimation over nine tissue parameters, producing more reliable diffusivity and T2 maps. In parallel, we developed a two-stage pipeline for diffusion-weighted breast MRI. The first stage uses self-supervised, physics-informed learning to estimate perfusion, diffusion, and blood-flow parameters without requiring expert-labeled ground truth. The second stage applies spatial refinement to suppress noise and artifacts while preserving clinically important tumor boundaries. Together, these approaches improve the stability, speed, and interpretability of quantitative MRI under conditions where traditional fitting methods fail, supporting more reliable cancer imaging and a practical path toward real-time, contrast-free tissue characterization at the scanner.
Watch the team present this project at 01:22:06 in the session recording here.
Keywords: Physics-informed deep learning, quantitative MRI, tissue parameter estimation, cancer imaging
Faculty Advisor
Batu Gundogdu holds a Ph.D. in Electrical Engineering and is an Associate Clinical Professor at the University of Chicago, bringing over 15 years of experience developing deep-learning models across the fields of speech, language, and medical imaging. Most recently, he served as Head of AI Engineering at Eva, an AI compute startup, building algorithms for a novel next-generation AI hardware.
Previously, Dr. Gundogdu served as Research Faculty in the Department of Radiology at the University of Chicago for four years, where he led AI efforts at the MRI Research Center.
In the UChicago MS in Applied Data Science program, he shares his practical insights and experiences with students on being an AI engineer and scientist across diverse venues, including research institutions (UChicago MRI Research Center), corporate companies (Analog Devices – ADI), government (Navy and NATO), and startups (Eva). He teaches courses on Bayesian Machine Learning for Generative AI and Deep Learning.
Dr. Gundogdu also teaches in UChicago’s MS in Applied Artificial Intelligence program, where he is a founding faculty member. Drawing on his experience across research, industry, and startups, he helps students connect advanced AI concepts with the realities of building and deploying AI systems in practice. He is also a Co-Founder of VicVox, a boutique AI startup analyzing social discourse.