Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Learn about groundbreaking research in medical AI through this conference talk that introduces CT-ScanGaze, the first publicly available eye gaze dataset for Computed Tomography (CT) imaging. Discover how understanding radiologists' eye movements during CT reading can enhance computer-aided diagnosis systems and explore the challenges of analyzing three-dimensional medical imaging data. Examine the novel CT-Searcher model, a 3D scanpath predictor specifically designed to process CT volumes and generate radiologist-like fixation sequences, addressing limitations of existing 2D-only approaches. Understand the innovative pipeline that converts 2D gaze datasets into 3D training data and explore comprehensive evaluation frameworks for 3D scanpath prediction in medical imaging. Gain insights into the intersection of expert behavior analysis and deep learning from PhD candidate Trong Thang Pham, whose research focuses on applying human eye gaze analysis to transform deep learning models, with publications in premier venues including CVPR, ICCV, and WACV. Participate in interactive discussion and Q&A as part of Stanford's MedAI Group Exchange Sessions, designed to critically examine key topics in AI and medicine while fostering collaborative learning among researchers and practitioners.
