Biostatistics and Bioinformatics Monthly Seminar with Michael R. Kosorok, PhD
Artificial Intelligence Driven Precision Health
Michael R. Kosorok, PhD
W.R. Kenan, Jr. Distinguished Professor, Department of Biostatistics
Professor, Department of Statistics and Operations Research
Director, Precision Health and Artificial Intelligence Research (PHAIR) Lab
Director, Center for Artificial Intelligence and Public Health
University of North Carolina at Chapel Hill
Meeting ID: 834 2878 7778
Passcode: Sy6FTVFi
Abstract:
I will be discussing some of the work we do at the UNC Center for AI and Public Health, including some recent applied work on using a SMART design combined with off-policy reinforcement learning to find a dynamic treatment regime (DTR) for treating chronic lower back pain. The approach we take is applicable to real-world healthcare challenges, where there are often multiple competing outcomes of interest, such as treatment efficacy and side effect severity. However, statistical methods for estimating dynamic treatment regimes usually assume a single outcome of interest, and the few methods that deal with composite outcomes suffer from important limitations. These include restrictions to a single time point and two outcomes, the inability to incorporate self-reported patient preferences and limited theoretical guarantees. To this end, we propose a new method to address these limitations, which extends SMART designs and Q-Learning to the composite outcome setting to achieve the ideal trade-off between outcomes for each patient. Our framework allows for an arbitrary number of time points and outcomes, incorporates stated preferences and achieves strong asymptotic performance with realistic assumptions on the data. We conduct simulation experiments based on an ongoing trial for chronic low back pain (BEST) as well as a well-known completed trial for schizophrenia. In all experiments, our method achieves highly competitive empirical performance compared to alternatives.
This group meets monthly from Noon – 1 p.m.