Courant Holiday Lecture: Imaging without Images: Using Artificial Intelligence for Direct Discovery of Spatial Signatures in the Absence of Image Reconstruction

Speaker: Sumit Chopra, New York University

Location: TBA

Date: Friday, December 12, 2025

Location: Warren Weaver Hall, Room TBA

For decades, fields like medical imaging, astronomy, and geoscience have relied on a central assumption that meaningful discoveries require high-fidelity images designed for human interpretation. While this image-first approach has driven significant advances, it also comes with steep costs. Producing such images requires large volumes of precise measurements, complex and expensive equipment, long acquisition times, and expert analysts - all of which limit who can access and benefit from these technologies. In this talk, using healthcare as a central example, I will show how AI allows us to rethink this paradigm entirely. By enabling the direct extraction of clinical insight without first producing human-readable images, AI has the potential to dramatically accelerate diagnosis, reduce costs, broaden access to care, and ultimately reshape how we deliver and experience healthcare.