Random Survival Forests

@article{Ishwaran2019RandomSF,
  title={Random Survival Forests},
  author={Hemant Ishwaran and Udaya B. Kogalur and Eugene H. Blackstone and Michael S. Lauer},
  journal={Wiley StatsRef: Statistics Reference Online},
  year={2019}
}
We introduce random survival forests, a random forests method for the analysis of right-censored survival data. New survival splitting rules for growing survival trees are introduced, as is a new missing data algorithm for imputing missing data. A conservation-of-events principle for survival forests is introduced and used to define ensemble mortality, a simple interpretable measure of mortality that can be used as a predicted outcome. Several illustrative examples are given, including a case… 

Figures from this paper

Random Survival Forests Random Survival Forests
Random survival forests (RSF) is a flexible nonparametric tree-ensemble method for the analysis of right-censored survival data. In this article, we provide a short overview of RSF. We review
Rotation survival forest for right censored data
TLDR
This research investigates the plausibility of extending a rotation forest, originally proposed for classification purpose, to survival analysis and shows that rotation survival forests are able to outperform the state-of-art survival ensembles on right censored data.
Survival Forests with R-Squared Splitting Rules
TLDR
By studying the well-known benchmark data sets, this work finds that the proposed model generally outperforms popular survival models such as random survival forest with different splitting rules, Cox proportional hazard model, and generalized boosted model in terms of C-index metric.
Analysis of Censored Data: Survival Trees and Random Forests
We elucidated in Chapter 4 the usefulness of the recursive partitioning technique for the analysis of binary outcomes. Not only can this technique be extended to the analysis of censored data, but
Doubly robust survival trees
TLDR
New 'doubly robust' extensions of these loss estimators motivated by semiparametric efficiency theory for missing data that better utilize available data are proposed.
Variable Selection with Random Survival Forest and Bayesian Additive Regression Tree for Survival Data
TLDR
A survival analysis methodology incorporating Bayesian additive regression trees to account for nonlinear and additive covariate effects and Predictive potential of the three models are compared using some widely used performance assessment measures in survival literature.
Adapting Random Forests to Cope with Heavily Censored Datasets in Survival Analysis
TLDR
A new Random Forest variant is proposed to cope with censored data, and evaluated in experiments predicting the time to diagnosis of 8 age-related diseases, for data from the English Longitudinal Study of Ageing (ELSA) database.
Survival forests for data with dependent censoring
TLDR
A new method for building survival forests, called p-forest, is proposed that may be used not only when dependent censoring is suspected, but also as a new survival forest method in general.
Survival Forests under Test
TLDR
Theoretical considerations explain the low power of log-rank-based splitting in detecting patterns in non-proportional hazards situations in survival trees and corresponding forests and can potentially be overcome by the alternative split procedures suggested herein.
Random rotation survival forest for high dimensional censored data
TLDR
It is shown that the proposed method random rotation survival forest outperforms state-of-the-art survival ensembles such as random survival forest and popular regularized Cox models in high dimensional censored time-to-event data analysis.
...
...

References

SHOWING 1-10 OF 106 REFERENCES
Survival Trees by Goodness of Split
Abstract A tree-based method for censored survival data is developed, based on maximizing the difference in survival between groups of patients represented by nodes in a binary tree. The method
Doubly robust survival trees
TLDR
New 'doubly robust' extensions of these loss estimators motivated by semiparametric efficiency theory for missing data that better utilize available data are proposed.
Relative risk trees for censored survival data.
TLDR
A method is developed for obtaining tree-structured relative risk estimates for censored survival data using a recursive partitioning algorithm that adopts most aspects of the widely used Classification and Regression Tree (CART) algorithm.
Random survival forests for competing risks.
TLDR
A new approach to competing risks using random forests is introduced and it is shown that the method is highly effective for both prediction and variable selection in high-dimensional problems and in settings such as HIV/AIDS that involve many competing risks.
Identifying Important Risk Factors for Survival in Patient With Systolic Heart Failure Using Random Survival Forests
TLDR
An RSF model in a cohort of patients with heart failure performed as well as a traditional Cox proportional hazard model and may serve as a more intuitive approach for clinicians to identify important risk factors for all-cause mortality.
Recursively Imputed Survival Trees
  • R. Zhu, M. Kosorok
  • Computer Science
    Journal of the American Statistical Association
  • 2012
TLDR
This new nonparametric regression procedure uses a novel recursive imputation approach combined with extremely randomized trees that allows significantly better use of censored data than previous tree-based methods, yielding improved model fit and reduced prediction error.
Survival ensembles.
TLDR
A unified and flexible framework for ensemble learning in the presence of censoring for right-censored data is proposed and a random forest algorithm and a generic gradient boosting algorithm are introduced for the construction of prognostic and diagnostic models.
Random Survival Forests.
  • Jeremy MG Taylor
  • Computer Science
    Journal of thoracic oncology : official publication of the International Association for the Study of Lung Cancer
  • 2011
TLDR
RSF is an adaptation of Random Forests designed to be used for survival data, and it has been shown that injecting some controlled variation or randomness into the construction of each of the separate trees can improve prediction performance.
High-Dimensional Variable Selection for Survival Data
TLDR
This work derives the distribution of the minimal depth and uses it for high-dimensional variable selection using random survival forests, and develops a new regularized algorithm, termed RSF-Variable Hunting.
Exponential survival trees.
TLDR
Simulation results demonstrate the recursive partitioning algorithm's capability to identify the underlying hazard structure in survival data and its integration into a computer program similar in structure to the commercial package CART.
...
...