Learning the Language of Patients - Multimodal Generative AI for Precision Health

Explore the intersection of multimodal generative AI and precision health in this comprehensive lecture by Hoifung Poon from Microsoft Research. Discover how technological advances and social policies have accelerated the digitization of multimodal, longitudinal patient data including electronic health records, medical imaging, and multiomics information. Learn about the development of powerful multimodal patient embeddings that serve as digital twins for patients, enabling the synthesis of complex medical information across millions of cancer patients. Examine how population-scale real-world evidence is being applied to advance precision oncology through partnerships with major health systems and life sciences companies. Gain insights into groundbreaking research on large language models and multimodal generative AI in healthcare applications, including the development of popular open-source foundation models such as PubMedBERT, BioGPT, BiomedCLIP, LLaVA-Med, and BiomedParse. Understand the practical applications of AI systems currently in daily use for molecular tumor boards and clinical trial matching, and learn about GigaPath, the first whole-slide digital pathology foundation model pretrained on over one billion pathology image tiles as featured in Nature.