TLDR: Jason Hu, a doctoral student in Electrical and Computer Engineering at the University of Michigan, has been granted the prestigious J. Robert Beyster Computational Innovation Graduate Fellowship. This award recognizes and supports his groundbreaking research in applying generative AI and machine learning to enhance imaging systems, particularly through novel diffusion models for high-resolution image reconstruction in medical and general applications.
Ann Arbor, MI – Jason Hu, a promising doctoral student in the Electrical and Computer Engineering (ECE) department at the University of Michigan, has been honored with the J. Robert Beyster Computational Innovation Graduate Fellowship. This significant award will provide crucial support for Hu’s cutting-edge research focused on developing generative artificial intelligence (AI) and machine learning (ML) algorithms for advanced image processing applications.
Hu’s research addresses the growing demand for more efficient imaging systems capable of producing higher-quality images with fewer resources. As Hu explains, “We live in a world where imaging systems are ubiquitous. From the cell phones in our pockets to our cars and doorbells and on to telescopes and medical scanners, imaging has changed how we share, document, and understand our world. There is an increasing demand to make these systems more efficient by producing higher-quality images with fewer and fewer resources. When we need an image that is of higher quality than can be directly constructed from its samples, we have an inverse problem and must rely on reconstruction schemes.”
To tackle these challenges, Hu has pioneered a novel diffusion model method designed to solve high-resolution 2D and 3D inverse problems. These problems would typically be computationally infeasible using traditional diffusion models. A key innovation in his approach is the ability of his models to solve out-of-distribution inverse problems, which significantly reduces the necessity for extensive training datasets. Furthermore, Hu has demonstrated that a single diffusion model can effectively solve inverse problems across multiple modalities, eliminating the need to train individual networks for every distinct dataset.
The practical applications of Hu’s research are diverse and impactful, extending to critical areas such as 3D computed tomography (CAT scans), magnetic resonance imaging (MRI), and generalized deblurring of images.
Professor Liyue Shen, who co-advises Hu alongside Jeff Fessler, the William L. Root Distinguished University Professor of Electrical Engineering and Computer Science, lauded his contributions. “Jason is conducting cutting-edge research that has made significant contributions in computational imaging for biomedical applications,” stated Prof. Shen. “His research has led to innovative methodology that expands the use of generative AI models.”
Professor Jeff Fessler also highlighted Hu’s exceptional academic trajectory, noting, “Jason was one of the very few brave students who took Image Processing (ECE 556) when he was still an undergraduate.” Hu’s academic excellence was evident early on; he earned the William J. Branstrom Freshman Prize for academic excellence after his first year at Michigan and was mastering graduate-level courses midway through his undergraduate career, where he also completed his bachelor’s degree in Electrical Engineering.
Also Read:
- Johns Hopkins Celebrates Groundbreaking Innovations in Fetal Surgery and AI-Driven Drug Discovery
- Tianyang Chen Receives Noble Business Award for Revolutionizing Design with AI
This fellowship underscores the University of Michigan’s commitment to fostering innovation in AI and its applications, particularly in the critical field of computational imaging.
