Responsible Use of Large Language Models in Healthcare

Speaker: Monica Agrawal

REGISTER: https://duke.qualtrics.com/jfe/form/SV_0OinoHO2fuBaSA6

Speaker Bio and Abstract

Speaker Biography: Dr. Monica Agrawal is an assistant professor at Duke University, jointly appointed between the Department of Biostatistics and Bioinformatics and the Department of Computer Science. Her research tackles diverse challenges including real-world evaluation of large language models in medicine, smarter electronic health records, and human-AI interaction. She has been named a Duke Whitehead Scholar, a Rising Star in EECS, and a finalist for the AMIA Doctoral Dissertation award. Dr. Agrawal earned her PhD in Computer Science at MIT in 2023 and is also a co-founder of Layer Health.

Abstract: Language is embedded across medicine, from clinical notes to medical literature to patient communication. Natural language processing, and particularly large language models, can help us to reimagine how we process and communicate this clinical text, which could have transformative effects for the practice of medicine. This talk will discuss open challenges, opportunities and solutions for NLP to accelerate clinical discovery for researchers, streamline workflows at the point-of-care for physicians, and improve the accessibility of health information for patients. First, I will discuss scalable techniques for clinical information extraction that leverage large language models. Next, I will describe a paradigm for smarter electronic health records that decreases documentation burden, incentivizes the creation of high-quality data at the point-of-care, and aids in proactive information retrieval. This will include a discussion of the difficulties of evaluating generative AI deployments in medicine. Finally, I will discuss our work studying how patients may be using large language models to access health information and advice, and the opportunities and pitfalls of doing so.