HI Seminar Series - Machine Learning in Biostatistics

September 6, 2017 - 4:00 PM to September 7, 2017 - 3:59 PM


Informatics Research Seminar: Machine Learning in Biostatistics

September 6 @ 4:00 pm - 5:00 pm

Speaker: Michael Kosorok, PhD
Presented from UNC-CH

Broadcast Link: Seminar

The Informatics Research Seminar Series is sponsored by Duke University and a collaboration with UNC-Chapel Hill, NCCU, UNC Charlotte, and ECU. This series explores key areas in Health Informatics and include research results, overview of programs of research, basic, applied, and evaluative projects, as well as research from varied epistemological stances.

An overview of how data science and big data have become relevant to biomedical research will be presented, including the roles of machine learning and statistics in data-driven decision making and precision medicine and the potential pitfalls of “big data hubris.” Several ongoing projects using machine learning for precision medicine will be described, including work on patient-derived xenograft mouse models for cancer, type 1 diabetes, and alcohol dependency. Sequential multiple assignment randomized trials (SMARTs) for precision medicine research will also be introduced.

Michael R. Kosorok, PhD, MS is the W.R. Kenan, Jr. Distinguished Professor and Chair of Biostatistics, Professor of Statistics and Operations Research, and Member of the Lineberger Comprehensive Cancer Center at the University of North Carolina at Chapel Hill. His research interests are in biostatistics, data science, machine learning, and precision medicine. He has authored over 130 peer-reviewed journal publications and two books, one on the theoretical foundations of biostatistics (Introduction to Empirical Processes and Semiparametric Inference, 2008, Springer) and one, co-authored with Erica E.M. Moodie, on current data science methods in precision medicine (Adaptive Treatment Strategies in Practice: Planning Trials and Analyzing Data for Personalized Medicine, 2016, ASA-SIAM). He is an honorary fellow of the American Statistical Association, the Institute of Mathematical Statistics, and the American Association for the Advancement of Science.