Data Sheet 1_Use of machine learning models to predict mortality in dialysis patients.pdf

BackgroundMortality among maintenance hemodialysis patients remains high, and traditional statistical models often fail to capture complex clinical relationships. This study aimed to systematically develop, compare, and validate 19 machine learning algorithms for predicting all-cause mortality in maintenance hemodialysis patients.
MethodsThis retrospective study included data from 538 maintenance hemodialysis patients (2018.1–2023.12), with 70% used for training and 30% for testing. Each model underwent hyperparameter optimization based on three performance metrics (accuracy, F1-score, and ROC Area Under the Curve [AUC]) to evaluate the impact of different clinical priorities.
ResultsGradient boosting models demonstrated consistent superiority, with performance outcomes highly sensitive to the selected optimization target. XGBoost optimized for accuracy achieved an F1 score of 0.683 and a ROC AUC of 0.899. AdaBoost optimized for F1 score attained the highest ROC AUC of 0.903 and an F1 score of 0.682. AdaBoost also demonstrated robust performance across optimization strategies, suggesting its suitability for clinical implementation where balanced risk prediction is essential.
ConclusionA systematic ML framework can yield tailored, high-performing models for mortality risk stratification in maintenance hemodialysis patients, with significant potential to enhance identification and management of high-risk individuals in clinical practice.
Clinical trial number RegistryChinese Clinica Trial Registry (ChiCTR), TRN:ChiCTR2500103960, Registration date: 9 June 2025.

背景：维持性血液透析（maintenance hemodialysis）患者的全因死亡率（all-cause mortality）仍处于较高水平，而传统统计模型往往难以捕捉复杂的临床关联关系。本研究旨在系统构建、对比并验证19种机器学习（Machine Learning）算法，以预测维持性血液透析患者的全因死亡率。
方法：本项回顾性研究纳入了2018年1月至2023年12月期间的538名维持性血液透析患者的数据，其中70%用于模型训练，剩余30%用于模型测试。所有模型均基于三项性能指标（准确率（accuracy）、F1分数（F1-score）、受试者工作特征曲线下面积（ROC Area Under the Curve, AUC））开展超参数优化（hyperparameter optimization），以评估不同临床优先级的影响。
结果：梯度提升模型展现出持续的性能优势，其预测结果对所选优化目标具有高度敏感性。以准确率为优化目标的XGBoost模型，其F1分数达0.683，ROC AUC为0.899。以F1分数为优化目标的AdaBoost模型，获得了最高的ROC AUC值0.903，F1分数为0.682。此外，AdaBoost模型在各类优化策略下均表现出稳健的性能，提示其适用于平衡风险预测至关重要的临床场景。
结论：系统性机器学习框架可针对维持性血液透析患者的死亡率风险分层构建定制化高性能模型，在临床实践中具备显著提升高危人群识别与管理水平的潜力。
临床试验注册号：本研究已在中国临床试验注册中心（Chinese Clinical Trial Registry, ChiCTR）完成注册，试验识别号（TRN）为ChiCTR2500103960，注册日期为2025年6月9日。