Biostatistics and Medical Informatics Seminar with Pengtao Xie (UCSD)

Semantic segmentation of medical images is pivotal in disease diagnosis and treatment planning. A major hurdle is the need for numerous research-intensive annotated masks, which often leads to ultra-low data regimes where annotated images are scarce, challenging the generalization of deep learning models on test images. Dr. Xie will introduce two complementary approaches: developing foundation models and generating high-fidelity training data of paired segmentation masks and medical images.