Greedy outcome weighted tree learning of optimal personalized treatment rules.

TitleGreedy outcome weighted tree learning of optimal personalized treatment rules.
Publication TypeJournal Article
Year of Publication2017
AuthorsZhu, Ruoqing, Ying-Qi Zhao, Guanhua Chen, Shuangge Ma, and Hongyu Zhao
JournalBiometrics
Volume73
Issue2
Pagination391-400
Date Published2017 06
ISSN1541-0420
KeywordsAlgorithms, Humans, Periodontics
Abstract

We propose a subgroup identification approach for inferring optimal and interpretable personalized treatment rules with high-dimensional covariates. Our approach is based on a two-step greedy tree algorithm to pursue signals in a high-dimensional space. In the first step, we transform the treatment selection problem into a weighted classification problem that can utilize tree-based methods. In the second step, we adopt a newly proposed tree-based method, known as reinforcement learning trees, to detect features involved in the optimal treatment rules and to construct binary splitting rules. The method is further extended to right censored survival data by using the accelerated failure time model and introducing double weighting to the classification trees. The performance of the proposed method is demonstrated via simulation studies, as well as analyses of the Cancer Cell Line Encyclopedia (CCLE) data and the Tamoxifen breast cancer data.

DOI10.1111/biom.12593
Alternate JournalBiometrics
Original PublicationGreedy outcome weighted tree learning of optimal personalized treatment rules.
PubMed ID27704531
PubMed Central IDPMC5378692
Grant ListR01 DK108073 / DK / NIDDK NIH HHS / United States
UL1 TR001863 / TR / NCATS NIH HHS / United States
R21 CA191383 / CA / NCI NIH HHS / United States
P01 CA154295 / CA / NCI NIH HHS / United States
P01 CA142538 / CA / NCI NIH HHS / United States
P30 CA016359 / CA / NCI NIH HHS / United States
U10 CA180819 / CA / NCI NIH HHS / United States
R01 GM059507 / GM / NIGMS NIH HHS / United States
P50 CA196530 / CA / NCI NIH HHS / United States