New framework enables reproducible benchmarking of competing risks survival models

Researchers have developed an open-source framework for benchmarking competing risks survival models, addressing a gap in systematic evaluation and adoption of these statistical and machine learning methods. The framework allows for comprehensive comparison of models across various datasets and performance metrics, including calibration, discrimination, prediction error, and clinical utility. Additionally, it introduces an extension of SHAP for competing risks, enabling model-agnostic interpretation of covariate contributions over time. The code is publicly available on GitHub. AI

IMPACT Provides a standardized approach for evaluating and comparing survival analysis models, potentially accelerating research and adoption in fields requiring such analysis.

RANK_REASON The item describes a new academic paper detailing a novel framework for statistical modeling. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv stat.ML →

• alphaXiv
• arXivLabs
• Begoña Sierra
• CatalyzeX Code Finder for Papers
• DagsHub
• GitHub
• Gotit.pub
• Hugging Face
• Influence Flower
• ScienceCast
• Shap

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →