Enhanced Survival Trees

2025-09-22

AI Summary

Traditional survival trees suffer from computational inefficiency, variable selection bias under censoring, redundant tree structure, poor interpretability, and unreliable confidence intervals for subgroup median survival times. To address these issues, we propose a novel survival tree method featuring: (i) an endpoint-bias-robust splitting criterion integrated with cross-validation to mitigate variable selection bias; (ii) a fusion-based regularization mechanism that merges statistically similar terminal nodes, enhancing structural parsimony and interpretability; and (iii) a pruning-assisted, bootstrap-based bias-corrected procedure for constructing valid confidence intervals for subgroup median survival times. Extensive simulations and analysis of the Alzheimer’s Disease Neuroimaging Initiative (ADNI) dataset demonstrate that the proposed method significantly improves predictive accuracy, model simplicity, and statistical inference reliability. It provides a computationally efficient and statistically rigorous tool for clinically meaningful subgroup identification in survival analysis.

Technology Category

• Machine Learning: Ensemble Methods
• Search and Optimization: Mixed Discrete/Continuous Search
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

We introduce a new survival tree method for censored failure time data that incorporates three key advancements over traditional approaches. First, we develop a more computationally efficient splitting procedure that effectively mitigates the end-cut preference problem, and we propose an intersected validation strategy to reduce the variable selection bias inherent in greedy searches. Second, we present a novel framework for determining tree structures through fused regularization. In combination with conventional pruning, this approach enables the merging of non-adjacent terminal nodes, producing more parsimonious and interpretable models. Third, we address inference by constructing valid confidence intervals for median survival times within the subgroups identified by the final tree. To achieve this, we apply bootstrap-based bias correction to standard errors. The proposed method is assessed through extensive simulation studies and illustrated with data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) study.

Problem

• Develops efficient survival trees for censored failure time data analysis
• Creates parsimonious tree structures using fused regularization and pruning
• Provides valid confidence intervals for median survival times in subgroups

Innovation

• Computationally efficient splitting with intersected validation
• Fused regularization framework for tree structure determination
• Bootstrap-based confidence intervals for median survival times