Abstract:
|
While the goal of personalized medicine-targeting treatments to take patient heterogeneity into account-has been a focus of medicine for centuries, recent years have seen an explosion of interest and activity in the field. In large part, this is due to the rise of precision medicine, which seeks to develop personalized medicine that is empirically based, scientifically rigorous, and reproducible. We present several new developments in statistical machine learning which assist in understanding patients who may differ in substantial ways. One of these is outcome-weighted learning, or O-learning, which directly estimates individualized treatment decision rules without requiring regression modeling and is thus robust to model misspecification. We will also describe several other new developments and illustrate these approaches with clinical examples, paying particular attention to dynamic biological systems that vary over time.
|