Personalized medicine and artificial intelligence
*Michael Rene Kosorok, University of North Carolina at Chapel Hill
Keywords: clinical trials, dynamic treatment regimes, machine learning, reinforcement learning
Personalized medicine has become an increasingly important area of medical research because of its potential to dramatically improve clinical outcome by giving the right treatment to the right person at the right time. Dynamic treatment regimens are specially designed sequences of treatments individualized to each patient to maximize clinical outcome. Each component treatment rule corresponds to a milestone in the disease and treatment process and maps up-to-date patient-level information to a recommended treatment. Statistics plays a key role in the construction of dynamic treatment regimens using observational and randomized clinical trial data. In this presentation, we show how tools from artificial intelligence can be utilized to find such individualized treatment rules. Examples based on treating depression and non-small cell lung cancer will presented, and both recent developments and open research questions will be discussed.
Important Dates & Deadlines
- October 9 - 11, 2013