Session Description

Generative AI is rapidly transforming the landscape of health care, offering novel solutions for clinical workflows, patient engagement, and research. However, evaluating its use cases requires careful consideration to ensure effectiveness and safety. In this session, we will discuss emerging trends as generative AI solutions proliferate the marketplace and how to best evaluate these solutions in terms of their effectiveness. We will discuss real world examples and how their evaluation can be used to not only ensure the safety of these tools, but also to improve them and mold them to custom clinical workflows.

Video

view slides

Speakers

Michael Gao, PhD

Dr. Michael Gao, (PhD Biostatistics, Duke University ’25) , serves as Principal AI Scientist at the Duke Institute for Health Innovation. He leads the development and technical implementation of machine learning and AI solutions in clinical care and health care operations, having deployed dozens of algorithmic tools such as real-time detection systems for sepsis, patient decompensation, and mortality risk. He has also designed and analyzed clinical trials to evaluate these interventions. His work has been published and presented in venues like AAAI, NeurIPS, JAMA Open, and NPJ Digital Medicine. His current work involves realizing the potential of generative AI tools in medicine and researching the human-computer interactions that result

other health AI hubs