BME Seminar: “Generative and Agentic AI in Healthcare”

Details

Quanzheng Li, Ph.D.
Director of Center for Advanced Medical Computing and Analysis, MGH,
Harvard Medical School

Abstract: Generative AI, which includes large language model (LLM), multimodality large model, and diffusion model, and Agentic AI has made tremendous progress in the last few years in computer science. In this talk, I will quick review generative AI models and then share our experience to adapt these technologies to various clinical applications. I will also briefly introduce clinical agentic AI, and demonstrate how we use agentic AI systems in real world clinical applications.

Bio: Quanzheng Li is the senior director for research and development, data science office, Massachusetts General Brigham, and the director of the Center for Advanced Medical Computing and Analysis, Massachusetts General Hospital, Harvard Medical School. He received his Ph.D degree in Electrical Engineering from the University of Southern California (USC) in 2005. His research interests include: (1) deep learning, particularly generative and agentic AI, on multimodality clinical data, including imaging, times series and EHR, for the screening, risk prediction, diagnosis, treatment optimization and prognosis of various diseases; (2) image reconstruction methods in PET, CT and MRI, especially in TOF and dynamic PET imaging (kinetic analysis); and (3) medical image analysis and machine learning with application to biomedical image registration of multi-modality longitudinal PET, MR and CT images. Dr. Li is the recipient of 2015 IEEE NPSS early achievement award, and 2024 Google Research Scholar. As the senior director at MGB data science office, Dr. Li overseen the data retrieval, curation, model development, testing, and deployment of 20+ AI models in various areas of medicine, and have worked with clinical champions in radiology, pathology, oncology, cardiology, surgery, and neurology to develop and deploy AI models in clinical settings.