Generative AI for Multimodal Biomedicine

Explore a comprehensive lecture on the advancement of generative AI in multimodal biomedicine presented by Assistant Professor Sheng Wang from the University of Washington's Paul G. Allen School of Computer Science & Engineering. Delve into three groundbreaking foundation models: GigaPath, the pioneering whole-slide pathology model for gigapixel-level images; OCTCube, the first 3D OCT retinal imaging model demonstrating superior performance across 27 tasks; and BiomedParse, an innovative multi-modal system integrating nine major biomedical imaging modalities. Learn how these models are revolutionizing medical applications through state-of-the-art performance in cancer detection, retinal disease prediction, and cross-modality analysis. Gain insights from Wang's extensive research background, which spans from Peking University to Stanford School of Medicine, and discover how his work in developing large-scale models has been adopted by major biomedical institutes including Mayo Clinic and UW Medicine. The 32-minute presentation concludes with a discussion on the future integration of multi-modal generative AI with multi-omics datasets and multi-agent frameworks.