Register Here: https://pitt.zoom.us/meeting/register/hr6kKbfgQneBAT4GKsQYJg

About this event: The rapid advancement of AI and NLP technologies has significantly transformed healthcare by enhanving clinical decision-making and predictive analytics. Despite these advancements, ethical concerns have emerged, especially regarding the risk of perpetuating or exacerbating existing health disparities. Recent research has shown that AI models used for clinical risk prediction can inadvertently reinforce inequities, highlighting the urgent need for the ethical integration of AI into health informatics. In this talk, Dr. Liu will outline core principles of ethical AI and introduce commonly used fairness metrics relevant to healthcare applications. He will demonstrate how these principles can be operationalized in health informatics research using both structured electronic health record data and unstructured clinical notes. The session will cover approaches for fairness analysis, bias mitigation techniques, and interpretability assessments of language models. The talk will conclude with a discussion of current challenges and opportunities for future research in responsible and equitable AI development in healthcare.

About the speaker: Feifan Liu, PhD, FAMIA

Dr. Feifan Liu is an Associate Professor of the Department of Population and Quantitative Health Sciences at the University of Massachusetts Chan Madical School and the founding director of the innovative AI for Health (iAI4Health) Lab. He is trained in computer science with expertise in natural language processing (NLP), machine (deep) learning, and AI risk predictive modeling. Dr. Liu and his team won first place in two international challenge tasks: Medical Visual Question Answering (2018) and Gene Mutation/Disease Relation Extraction (2019). His research recently focuses on responsible AI for advancing health equity. He was awarded as an NIH AIM-AHEAD leadership fellow in 2022 and has been selected to serve as an AI/ML mentor for the NIH All of US research scholar program, and several NIH AIM-AHEAD training programs.