"Extracting Clinical Information from Millions of Pixels" by Mert R. Sabuncu, PhD

Abstract

Each radiology study contains millions of pixels, yet only a fraction of that information is ever used. In the United States, more than 100 million CT, MRI, and PET scans are performed every year. Radiologists interpret these data and summarize their findings in brief text reports that guide clinical decisions. But the underlying images—rich with quantitative, spatial, and temporal patterns—remain largely unexplored after the initial read. Artificial intelligence now offers a path to unlock this hidden information. At Cornell, Dr. Sabuncu's group develops methods that extract clinically meaningful insights directly from imaging data. In this talk, Dr. Sabuncu will highlight several recent projects shaped by real-world clinical needs. These include tracking disease progression over time, enabling scalable cancer screening in an opportunistic manner, and deriving quantitative measurements that can be integrated into routine care. He will also describe a simple framework for handling missing or incomplete data, a pervasive challenge in clinical imaging. Together, these efforts illustrate how AI can help radiology evolve from qualitative description to quantitative science—enhancing diagnosis, personalizing treatment, and transforming how we use medical images to improve patient care.

Bio

Mert R. Sabuncu received a PhD degree in Electrical Engineering from Princeton University, where his dissertation focused on entropy-based approaches to image registration. Mert then moved to the Massachusetts Institute of Technology for a post-doc with Polina Golland at the Computer Science and Artificial Intelligence Lab, where he worked on a range of biomedical image analysis problems, including the segmentation of brain MRI scans. After his post-doc at MIT, Mert spent a few years at the A.A Martinos Center for Biomedical Imaging (Massachusetts General Hospital and Harvard Medical School) as a junior faculty member, where he built a research program on algorithmic tools for integrating genetics and medical imaging. Today, Mert is a Professor in Electrical and Computer Engineering at Cornell University and Cornell Tech, in New York City. He also holds a dual appointment in Radiology at Weill Cornell Medicine, where he serves as the Vice Chair of AI and Engineering Research. His group develops machine learning based computational tools for biomedical imaging applications. He is a recipient of an NSF CAREER Award (2018) and an NIH Early Career Development Award (2011).