In a 1941 lecture, Dr. W. Edward Chamberlain was among the first to illustrate how our understanding of radiological interpretation could greatly benefit from knowledge about how the eye and brain process the light falling on the retina. Since then, medical image perception—a subfield that has attracted physicists, statisticians, engineers, and psychologists—has contributed to understanding radiological errors, providing tools to quantify observer performance, and developing computational models that are now routinely used to evaluate the task-based quality of medical images. In this talk, I will highlight some historical contributions, discuss present perceptual challenges regarding 3D imaging modalities and their solutions, and make the case for why understanding radiologists’ perceptual and decision capabilities remains essential in the age of AI-assisted reading.

Presented by Miguel Eckstein

Miguel Eckstein is the Duncan and Suzanne Mellichamp Professor in Mind and Machine Intelligence, Distinguished Professor of Psychological and Brain Sciences, and an Affiliate faculty member in the Departments of Computer Science and of Electrical and Computer Engineering, all at the University of California, Santa Barbara. He earned a Bachelor’s in Physics and Psychology at UC Berkeley and a PhD in Cognitive Psychology at UCLA. He then worked at the Department of Medical Physics and Imaging, Cedars Sinai Medical Center, and NASA Ames Research Center before moving to UC Santa Barbara. He is the recipient of the Optical Society of America Young Investigator Award, the National Science Foundation CAREER Award, the Anne and Donald Herbert Distinguished Lectureship on Modern Statistical Modeling, the National Academy of Sciences Troland Award, and a Guggenheim Fellowship. He served as the chair of the Vision Technical Group of the Optical Society of America, chair of the Human Performance, Image Perception and Technology Assessment conference of the SPIE Medical Imaging Annual Meeting, and as a member of various National Institute of Health study section panels over 35 times. He also held positions as the Vision Editor of the Journal of the Optical Society of America A, the board of editors of the Journal of Vision, the board of directors of the Vision Sciences Society, and Chair of the Gordon Conference. His research focuses on understanding how human vision and the perception of medical images. The research combines computational modeling, AI, behavioral studies, eye tracking, medical imaging, and neuroimaging. His work appears in a wide range of journals, including Proceedings of the National Academy of Sciences, Nature Communications, Nature Human Behavior, Current Biology, Journal of Neuroscience, Radiology, PLOS Computational Biology, IEEE Transactions in Medical Imaging, Computer Vision & Pattern Recognition (CVPR), etc.