Machine learning models did not significantly outperform conventional statistical models in predicting cardiovascular events in dialysis patients (mean AUC 0.784 vs 0.772, p=0.24).
Systematic Review (n=29,310)
Do machine learning models improve the prediction of cardiovascular events in dialysis patients compared to conventional statistical models?
Machine learning models overall do not significantly outperform conventional statistical models for predicting cardiovascular events in dialysis patients, though deep learning approaches show promise.
Absolute Event Rate: 0.784% vs 0.772%
p-value: p=0.24
= 0.727). Studies were predominantly originated from China (71.40%) and relied on internal validation (78.57%), limiting generalizability. Although deep learning algorithms show promises, ML models overall do not significantly outperform CSMs. CSMs remain viable, especially in resource-limited settings. Critical limitations include geographical bias, insufficient external validation, and tradeoffs between accuracy and interpretability. Future research should prioritize validation frameworks and clinical implementation over marginal accuracy improvements.
Lü et al. (Wed,) conducted a systematic review in Cardiovascular events in dialysis patients (n=29,310). Machine learning models vs. Conventional statistical models was evaluated on Model discrimination performance (AUC/C-index) (p=0.24). Machine learning models did not significantly outperform conventional statistical models in predicting cardiovascular events in dialysis patients (mean AUC 0.784 vs 0.772, p=0.24).
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: