Health Disparities From Epidemiological and Clinical Perspectives
Yuan Luo
February 8, 2023, Wednesday, 3:00 PM - 4:00 PM EST
I will discuss that ML models trained on datasets that lack demographic diversity could potentially yield sub-optimal performance when being applied to the underrepresented populations (ethnic minorities, lower social-economic status), thus perpetuating health disparity. Drawing from experiences of investigating this issue in both EHR datasets and epidemiological datasets, I will talk about disparities in ML model performances in racial, gender, and insurance subgroups. I will also discuss approaches to design simulation models that can expose unintended biases in ventilator allocation in critically ill patients with COVID-19.
Dr. Luo is currently Associate Professor at Department of Preventive Medicine, at Feinberg School of Medicine in Northwestern University. He is Chief AI Officer at Clinical and Translational Sciences Institute (NUCATS) and Institute for Augmented Intelligence in Medicine. Dr. Luo earned his PhD degree from MIT EECS with a math minor. He won the prestigious American Medical Informatics Association (AMIA) New Investigator Award in 2020. He served on AMIA Membership and Outreach Committee. He has published over 100 peer-reviewed papers. His publications appear in leading journals including Nature Medicine, JAMA, AJRCCM, JAMIA, JBI etc. He has published in and/or served as PC members for top AI and informatics conferences including AAAI, KDD, CVPR, ACL, AMIA etc.