Key result
Machine-learning models outperform the Cox model in predicting survival and recurrent MR.
Why the study?
This study was conducted to assess long-term clinical outcomes after mitral valve repair using machine-learning techniques.
Do machine-learning techniques improve the prediction of survival and recurrent mitral regurgitation in patients undergoing mitral valve repair compared to traditional Cox models?
Cohort (n=436)
Do machine-learning techniques improve the prediction of survival and recurrent mitral regurgitation in patients undergoing mitral valve repair compared to traditional Cox models?
Effect estimate: C-index 0.814 (RSF) and 0.806 (XGBoost)
Machine-learning models, particularly XGBoost and Random Survival Forest, provide superior predictive accuracy for long-term survival and recurrent mitral regurgitation after mitral valve repair compared to traditional Cox proportional hazards models.
No takes yet. Share an insight, caveat, or question.
Machine-learning models may improve post-repair survival prediction; leaves open need for external validation before clinical use.
Kang et al. (2023) conducted a cohort in Mitral regurgitation (n=436). Machine-learning techniques (XGBoost, RSF) vs. Cox proportional hazards model was evaluated on Actuarial survival and freedom from significant (≥ moderate) mitral regurgitation (C-index 0.814 (RSF) and 0.806 (XGBoost)). Machine-learning models (XGBoost and random survival forest) outperformed the Cox model in predicting overall survival (C-index 0.806 and 0.814 vs 0.733) and recurrent mitral regurgitation.
Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context: