A reproducible and extensible framework for benchmarking competing risks survival models

📅 2026-07-31
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
This work addresses the lack of a systematic, reproducible, and scalable evaluation benchmark for competing risks survival analysis. It introduces the first open-source benchmarking framework specifically designed for competing risks models, enabling comprehensive assessment across multiple datasets along key dimensions including calibration, discrimination, overall prediction error, and clinical utility. The framework further advances model interpretability by innovatively extending SHAP (SHapley Additive exPlanations) to support time-dynamic, model-agnostic explanations. Compatible with a wide range of modeling approaches and accompanied by publicly released code, this benchmark significantly promotes standardized evaluation and interpretable deployment of competing risks models in practice.
📝 Abstract
A wide range of statistical and machine learning methods have been proposed for survival analysis with competing risks, where the occurrence of one event (i.e., cancer death) precludes the occurrence of other events (i.e., cardiovascular disease death). Despite these methodological advances, their systematic evaluation and adoption are limited by the lack of comprehensive, reproducible and extensible benchmarking frameworks. We developed an open-source benchmarking framework for competing risks models that enables their systematic comparison across multiple datasets under different aspects of performance; calibration, discrimination, overall prediction error and clinical utility. We additionally introduce an extension of SHAP for competing risks, allowing model-agnostic interpretability of covariates contributions over time. All our code is publicly available via GitHub:https://github.com/BBolosSierra/CompRisksBenchmark
Problem

Research questions and friction points this paper is trying to address.

competing risks
survival analysis
benchmarking
reproducibility
model evaluation
Innovation

Methods, ideas, or system contributions that make the work stand out.

competing risks
benchmarking framework
survival analysis
SHAP
model interpretability
🔎 Similar Papers
B
Begoña B. Sierra
Cancer Research UK Scotland Centre, Institute of Genetics and Cancer, University of Edinburgh
C
Colin McLean
Cancer Research UK Scotland Centre, Institute of Genetics and Cancer, University of Edinburgh
P
Peter S. Hall
Cancer Research UK Scotland Centre, Institute of Genetics and Cancer, University of Edinburgh
S
Sarah Friedrich-Welz
Mathematical Statistics and AI in Medicine, University of Augsburg
C
Catalina A. Vallejos
Institute of Genetics and Cancer, University of Edinburgh