University of Michigan
105 Jordan Hall
Refreshments will be provided from 3:30 - 4:00 PM in the Jordan Hall Galleria
Supervised learning of health-related secret codes from wearable device data
Wearable devices are becoming a popular mini-robot to collect real-time information on well-being from a device user. One primary goal is to extract and validate features relevant to personal health conditions from massive high-frequency time-series measurements, so that each device user can utilize personal secret codes to guide personal health management. In this talk, I will introduce three smart personal wearables, Empatica E4, ActiGraph GTX and YONO Earbud, that have been used in various public health research projects, and then demonstrate how data collected from these devices may be used to solve some important health-related questions such as sleep health, physical activities, and reproductive health. I will give an overview on the application of different supervised learning techniques to process wearable device data and build tailored AI-systems to assist people with health-related decision making.
Bio: Dr. Song is Professor of Biostatistics at the Department of Biostatistics, School of Public Health in the University of Michigan, Ann Arbor. He received his PhD in Statistics from the University of British Columbia, Vancouver, Canada in 1996. He is a leading researcher in the field of correlated data analysis, has published over 190 peer- reviewed papers, and trained 22 PhD students and 5 postdocs. Dr. Song's current research interests include data integration, distributed inference, high-dimensional data analysis, longitudinal data analysis, mediation analysis, spatiotemporal modeling, and smart precision health. He collaborates extensively with researchers from nutritional sciences, environmental health sciences, chronic diseases, and nephrology. He is IMS Fellow, ASA Fellow and Elected Member of the International Statistical Institute. Dr. Song now serves as Associate Editor of the Journal of American Statistical Association, the Canadian Journal of Statistics, and the Journal of Multivariate Analysis.