AIMI Grand Rounds: The Hidden Layer: Radiologist–AI Interaction - Amy Hong, MD, PhD

The AIMI Grand Rounds, sponsored by the Center for Artificial Intelligence in Medicine and Imaging (AIMI), is a virtual series held on every fourth Tuesday. The series is a crucial initiative for disseminating the latest AI advancements in medicine, aiming to drive transformative innovations in healthcare. Stanford participants of the live event can claim 1.0 CME credits: AMA PRA Category 1 Credits ™ or Non-Physician Participation Credit.

Speaker:

Amy Hong, MD, PhD: Assistant Professor of Radiology, Stanford University 

Bio: Eun Kyoung (Amy) Hong, MD, PhD is a thoracic radiologist and physician-scientist at Stanford Radiology with a research focus on clinical evaluation of artificial intelligence in chest imaging. Her work centers on radiologist–AI interaction, generative AI for radiologic reporting, and the downstream clinical and workflow implications of AI deployment. She has led multi-institutional studies examining automation bias and variability in AI-assisted reporting across chest radiography and CT. Her research bridges technical AI performance with clinical deployment realities.

Abstract: Advances in vision and vision-language models have led to improved benchmark performance for a range of radiologic tasks. This talk will examine radiologist–AI interaction as a component of the overall system, with a focus on how model outputs are interpreted and used by clinicians in practice. Using examples from chest radiography and CT reader studies, the session will explore how factors such as output representation, uncertainty, and error characteristics may influence interpretation and reporting. The talk will also discuss considerations for evaluating AI systems beyond standalone model performance.