基于临床与心脏磁共振衍生特征对射血分数保留的心力衰竭进行多模态风险分层:一种纳入模型可解释性的方法
本研究为回顾性队列,纳入301例受试者(HFpEF组171例、对照组130例),整合临床变量与心脏磁共振衍生的心外膜脂肪组织(EAT)容积等特征构建可解释机器学习模型。HFpEF患者EAT容积(70.9±27.3 vs. 41.9±18.3 mL,p<0.001)和NT-proBNP水平显著更高;投票分类器诊断HFpEF准确率0.94,AdaBoost、XGBoost和随机森林预测再住院与全因死亡等不良结局准确率0.95。
要点
- HFpEF患者EAT容积为70.9±27.3 mL,显著高于对照组的41.9±18.3 mL(p<0.001)。
- 投票分类器诊断HFpEF的准确率为0.94、精确率0.96、召回率0.94、F1分数0.95。
- AdaBoost、XGBoost和随机森林预测再住院与全因死亡的准确率为0.95。
核心结论
- 结合临床与CMR衍生特征(尤其是EAT容积)的可解释模型可改善对HFpEF诊断和结局预测的支持,凸显数据驱动且可解释的方法在刻画HFpEF表型方面的价值,并可能促进特定人群的个体化风险评估。
研究设计与样本
- 回顾性队列研究,共301例受试者(HFpEF组171例、对照组130例),整合基线特征、CMR衍生的EAT容积和实验室生物标志物构建机器学习模型,采用准确率、精确率、召回率、F1分数、ROC-AUC和PR-AUC评估模型性能,并应用超参数优化和集成技术。
诊断标准与关键数值
- HFpEF组EAT容积70.9±27.3 mL vs. 对照组41.9±18.3 mL,p<0.001。
- HFpEF组NT-proBNP 1574 [963,2722] pg/mL vs. 对照组33 [10,100] pg/mL,p<0.001。
- 投票分类器诊断HFpEF准确率0.94、精确率0.96、召回率0.94、F1分数0.95。
- AdaBoost、XGBoost和随机森林预测再住院和全因死亡等不良临床结局准确率0.95。
临床应用与阅片要点
- 作者认为,结合临床与CMR衍生特征(尤其是EAT容积)的可解释模型可改善对HFpEF诊断和结局预测的支持。
- 作者认为,数据驱动且可解释的方法可能促进特定人群的个体化风险评估。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
英文原文(摘要)· 点击展开
RATIONALE AND OBJECTIVES: Heart failure with preserved ejection fraction (HFpEF) poses significant diagnostic and prognostic challenges due to its clinical heterogeneity. This study proposes a multi-modal, explainable machine learning framework that integrates clinical variables and cardiac magnetic resonance (CMR)-derived features, particularly epicardial adipose tissue (EAT) volume, to improve risk stratification and outcome prediction in patients with HFpEF.
MATERIALS AND METHODS: A retrospective cohort of 301 participants (171 in the HFpEF group and 130 in the control group) was analyzed. Baseline characteristics, CMR-derived EAT volume, and laboratory biomarkers were integrated into machine learning models. Model performance was evaluated using accuracy, precision, recall, and F1-score. Additionally, receiver operating characteristic area under the curve (ROC-AUC) and precision-recall area under the curve (PR-AUC) were employed to assess discriminative power across varying decision thresholds. Hyperparameter optimization and ensemble techniques were applied to enhance predictive performance.
RESULTS: HFpEF patients exhibited significantly higher EAT volume (70.9±27.3 vs. 41.9±18.3 mL, p<0.001) and NT-proBNP levels (1574 [963,2722] vs. 33 [10,100] pg/mL, p<0.001), along with a greater prevalence of comorbidities. The voting classifier demonstrated the highest accuracy for HFpEF diagnosis (0.94), with a precision of 0.96, recall of 0.94, and an F1-score of 0.95. For prognostic tasks, AdaBoost, XGBoost and Random Forest yielded superior performance in predicting adverse clinical outcomes, including rehospitalization and all-cause mortality (accuracy: 0.95). Key predictive features identified included EAT volume, right atrioventricular groove (Right AVG), tricuspid regurgitation velocity (TRV), and metabolic syndrome.
CONCLUSION: Explainable models combining clinical and CMR-derived features, especially EAT volume, improve support for HFpEF diagnosis and outcome prediction. These findings highlight the value of a data-driven, interpretable approach to characterizing HFpEF phenotypes and may facilitate individualized risk assessment in selected populations.
引用
- Zhang S, Lin Y, Han D, et al. Multi-modal Risk Stratification in Heart Failure with Preserved Ejection Fraction Using Clinical and CMR-derived Features: An Approach Incorporating Model Explainability. Acad Radiol. 2025. PMID: 40681365. doi:10.1016/j.acra.2025.06.048
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。