|
Activity Number:
|
112
- Statistical Challenges in the Processing and Analysis of Mobile Health Data
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, July 29, 2019 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #300248
|
Presentation
|
|
Title:
|
Precision Medicine in Mobile Health Using V-Learning
|
|
Author(s):
|
Daniel Luckett and Eric B Laber and Anna Kahkoska and David Maahs and Elizabeth Mayer-Davis and Michael Kosorok*
|
|
Companies:
|
University of North Carolina at Chapel Hill and NC State University and University of North Carolina at Chapel Hill and Stanford University and University of North Carolina at Chapel Hill and University of North Carolina at Chapel Hill
|
|
Keywords:
|
Markov decision processes;
Precision medicine;
Reinforcement learning;
Type 1 diabetes
|
|
Abstract:
|
We propose a new reinforcement learning tool for estimating optimal treatment regimes from mobile health data in real time. The approach is distinguished from other learning methods by modeling the value function as a direct function of the treatment regime. We establish consistency and asymptotic normality as well as compare to alternative approaches via simulations studies. We also describe an interesting application to type 1 diabetes.
|
Authors who are presenting talks have a * after their name.
