EE Colloquia: Model based deep learning (MoDL) for MR image recovery: beyond algorithm unrolling

Abstract: MRI is widely used as a medical imaging modality due to its superior soft-tissue contrast. Compressed sensing algorithms have been extensively employed in MRI to overcome the challenges associated with the slow nature of MRI acquisition. These methods offer guaranteed uniqueness, fast convergence, and stability properties. Model-based deep learning methods that combine imaging physics with learned regularization priors have emerged as more powerful alternatives for MR image recovery in recent years. The talk will introduce different flavors of model-based deep learning methods and discuss the unique challenges associated with these schemes in high-dimensional settings. Novel iterative algorithms that possess guarantees similar to compressive sensing while offering improved performance will be introduced.

Bio:  Mathews Jacob is a professor in the Department of Electrical and Computer Engineering and is heading the Computational Biomedical Imaging Group (CBIG) at the University of Iowa. His research interests include image reconstruction, image analysis, and quantification in the context of magnetic resonance imaging. He received his Ph.D. degree from the Biomedical Imaging Group at the Swiss Federal Institute of Technology in 2003. He was a Beckman postdoctoral fellow at the University of Illinois at Urbana Champaign. Dr. Jacob is the recipient of the CAREER award from the National Science Foundation in 2009, the Research Scholar Award from American Cancer Society in 2011, and the Faculty Excellence Award for Research from University of Iowa in 2021. He is currently the associate editor of the IEEE Transactions on Medical Imaging and has served as the associate editor of IEEE Transactions on Computational Imaging from 2016-20. He was the senior author on two best paper awards and one best machine learning paper award from IEEE ISBI. He was the general chair of IEEE International Symposium on Biomedical Imaging, 2020. He was elected as a Fellow of the IEEE (2022) for contributions to computational biomedical imaging.