Alomari Laith, Jarrar Yaman, Al-Fakhouri Zaid, Otabor Emmanuel, Lam Justin, Alomari Jana
Department of Medicine, Jefferson Einstein Philadelphia Hospital, Thomas Jefferson University, Philadelphia, Pennsylvania.
Department of Medicine, Lehigh Valley Health Network, Allentown, Pennsylvania.
Heart Rhythm O2. 2025 Feb 21;6(5):652-660. doi: 10.1016/j.hroo.2025.02.008. eCollection 2025 May.
Atrial fibrillation (AF) in critically ill patients increases morbidity, hospital stays, and costs. Existing prediction tools are limited in intensive care unit (ICU) settings.
This study developed a machine learning-based model to enable early AF risk identification and prevention.
In this retrospective cohort study, adult patients admitted to the ICU were identified from the MIMIC-IV (Medical Information Mart for Intensive Care-IV) database, including 47 clinical and laboratory variables. The primary outcome was AF within the first 48 hours of admission. Multiple machine learning models were trained to predict AF, with the top-performing model undergoing hyperparameter tuning. A compact model was developed using 15 variables and 2 novel features-one identifying patients 70 years of age or older with sepsis and another representing a composite score of pre-existing cardiac risk factors. Model performance was evaluated using accuracy, area under the receiver-operating characteristic curve (AUROC), and predictive values. SHAP (Shapley Additive exPlanations) analysis interpreted individual feature contributions to the model's predictions.
The cohort comprised 46,266 ICU patients, with 4.6% developing AF within 48 hours. The CatBoost classifier model achieved an AUROC of 0.850 on the test set, while the compact model with new features yielded an AUROC of 0.820. SHAP analysis highlighted total serum magnesium, age, and the newly created features as key predictors of AF development.
This study demonstrates the potential of machine learning models in predicting AF development in ICU patients. The compact model, with a satisfactory AUROC, can be a valuable tool for identifying high-risk patients and facilitating timely interventions.
重症患者的心房颤动（AF）会增加发病率、住院时间和费用。现有的预测工具在重症监护病房（ICU）环境中存在局限性。
本研究开发了一种基于机器学习的模型，以实现早期AF风险识别和预防。
在这项回顾性队列研究中，从MIMIC-IV（重症监护医学信息集市-IV）数据库中识别出入住ICU的成年患者，包括47个临床和实验室变量。主要结局是入院后48小时内发生AF。训练了多个机器学习模型来预测AF，性能最佳的模型进行超参数调整。使用15个变量和2个新特征开发了一个精简模型——一个用于识别70岁及以上患有脓毒症的患者，另一个代表既往心脏危险因素的综合评分。使用准确率、受试者操作特征曲线下面积（AUROC）和预测值评估模型性能。SHAP（Shapley加性解释）分析解释了各个特征对模型预测的贡献。
该队列包括46266名ICU患者，其中4.6%在48小时内发生AF。CatBoost分类器模型在测试集上的AUROC为0.850，而具有新特征的精简模型的AUROC为0.820。SHAP分析突出了血清总镁、年龄和新创建的特征是AF发生的关键预测因素。
本研究证明了机器学习模型在预测ICU患者AF发生方面的潜力。具有令人满意的AUROC的精简模型可以成为识别高危患者和促进及时干预的有价值工具。
