Nonparametric Failure Time: Time-to-Event Machine Learning with Heteroskedastic Bayesian Additive Regression Trees and Low Information Omnibus Dirichlet Process Mixtures
Biometrics · 2023
DOI: 10.1111/biom.13857 ↗Kurz gefasst
Fachartikel in der Zeitschrift Biometrics, 2023. Verfasst von Rodney A Sparapani und 4 weiteren Autor:innen. Beteiligte Einrichtung: Division of Biostatistics, Medical College of Wisconsin, Milwaukee, Wisconsin, USA. Hauptthemen laut MEDLINE: Software, Machine Learning. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 43 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NCI NIH HHS, NHLBI NIH HHS, Office of Naval Research.
Abstract (Originalsprache)
Many popular survival models rely on restrictive parametric, or semiparametric, assumptions that could provide erroneous predictions when the effects of covariates are complex. Modern advances in computational hardware have led to an increasing interest in flexible Bayesian nonparametric methods for time-to-event data such as Bayesian additive regression trees (BART). We propose a novel approach that we call nonparametric failure time (NFT) BART in order to increase the flexibility beyond accelerated failure time (AFT) and proportional hazard models. NFT BART has three key features: (1) a BART prior for the mean function of the event time logarithm; (2) a heteroskedastic BART prior to deduce a covariate-dependent variance function; and (3) a flexible nonparametric error distribution using Dirichlet process mixtures (DPM). Our proposed approach widens the scope of hazard shapes including nonproportional hazards, can be scaled up to large sample sizes, naturally provides estimates of uncertainty via the posterior and can be seamlessly employed for variable selection. We provide convenient, user-friendly, computer software that is freely available as a reference implementation. Simulations demonstrate that NFT BART maintains excellent performance for survival prediction especially when AFT assumptions are violated by heteroskedasticity. We illustrate the proposed approach on a study examining predictors for mortality risk in patients undergoing hematopoietic stem cell transplant (HSCT) for blood-borne cancer, where heteroskedasticity and nonproportional hazards are likely present.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Division of Biostatistics, Medical College of Wisconsin, Milwaukee, Wisconsin, USA.
- National Marrow Donor Program, Minneapolis, Minnesota, USA.
- School of Mathematical and Statistical Sciences, Arizona State University, Tempe, Arizona, USA.
ORCID-Profile: Rodney A Sparapani, Brent R Logan
Förderung
- NCI NIH HHS
- NHLBI NIH HHS
- Office of Naval Research
- National Cancer Institute
- National Heart, Lung, and Blood Institute
- National Institute of Allergy and Infectious Diseases
Zitieren
APA
Sparapani, R. A., Logan, B. R., Maiers, M. J., Laud, P. W., & McCulloch, R. E. (2023). Nonparametric Failure Time: Time-to-Event Machine Learning with Heteroskedastic Bayesian Additive Regression Trees and Low Information Omnibus Dirichlet Process Mixtures. Biometrics. https://doi.org/10.1111/biom.13857
Vancouver
Sparapani RA, Logan BR, Maiers MJ, Laud PW, McCulloch RE. Nonparametric Failure Time: Time-to-Event Machine Learning with Heteroskedastic Bayesian Additive Regression Trees and Low Information Omnibus Dirichlet Process Mixtures. Biometrics. 2023. doi:10.1111/biom.13857
BibTeX
@article{Sparapani2023,
author = {Sparapani, Rodney A and Logan, Brent R and Maiers, Martin J and Laud, Purushottam W and McCulloch, Robert E},
title = {Nonparametric Failure Time: Time-to-Event Machine Learning with Heteroskedastic Bayesian Additive Regression Trees and Low Information Omnibus Dirichlet Process Mixtures},
journal = {Biometrics},
year = {2023},
doi = {10.1111/biom.13857}
}
EUVIMED