Realising The Promise of Foundation Models in Healthcare
Abstract Large Language models like ChatGPT have captured the imagination of machine learning practitioners with their potential to transform the application of AI across many fields. However, in healthcare, transitioning from impressive tech demos to deployed AI has been challenging.
In this talk, I will discuss the opportunities large language models and other medical foundation models offer in terms of providing a better paradigm of doing “AI in healthcare.” First, I’ll outline what foundation models are and their relevance to healthcare. Then I’ll highlight some key opportunities provided by the next generation of medical foundation models. Finally I will discuss the current limitations in benchmarking and evaluating foundation models for medicine and how we can do better moving forward.
Speaker: Jason Alan Fries
Jason Fries is a computer scientist at Stanford University’s Center for Biomedical Informatics Research. His work focuses on methods that enable domain experts to rapidly build and modify machine learning models in complex domains such as medicine where obtaining large-scale, expert-labeled training data is a significant challenge. His research focuses on weakly supervised machine learning, foundation models, and methods for data-centric AI.