Liu Zheng, Shu Wenqi, Liu Hongyan, Zhang Xuan, Chong Wei
Department of Emergency, The First Hospital of China Medical University, Shenyang, China.
PLoS One. 2025 Feb 18;20(2):e0317819. doi: 10.1371/journal.pone.0317819. eCollection 2025.
Developing and validating interpretable machine learning (ML) models for predicting whether triaged patients need to be admitted to the intensive care unit (ICU).
The study analyzed 189,167 emergency patients from the Medical Information Mart for Intensive Care IV database, with the outcome being ICU admission. Three models were compared: Model 1 based on Emergency Severity Index (ESI), Model 2 on vital signs, and Model 3 on vital signs, demographic characteristics, medical history, and chief complaints. Nine ML algorithms were employed. The area under the receiver operating characteristic curve (AUC), F1 Score, Positive Predictive Value, Negative Predictive Value, Brier score, calibration curves, and decision curves analysis were used to evaluate the performance of the models. SHapley Additive exPlanations was used for explaining ML models.
The AUC of Model 3 was superior to that of Model 1 and Model 2. In Model 3, the top four algorithms with the highest AUC were Gradient Boosting (0.81), Logistic Regression (0.81), naive Bayes (0.80), and Random Forest (0.80). Upon further comparison of the four algorithms, Gradient Boosting was slightly superior to Random Forest and Logistic Regression, while naive Bayes performed the worst.
This study developed an interpretable ML triage model using vital signs, demographics, medical history, and chief complaints, proving more effective than traditional models in predicting ICU admission. Interpretable ML aids clinical decisions during triage.
开发并验证可解释的机器学习（ML）模型，用于预测分诊患者是否需要入住重症监护病房（ICU）。
该研究分析了重症监护医学信息数据库IV中的189,167例急诊患者，结局为入住ICU。比较了三种模型：基于急诊严重程度指数（ESI）的模型1、基于生命体征的模型2以及基于生命体征、人口统计学特征、病史和主要症状的模型3。采用了九种ML算法。使用受试者工作特征曲线下面积（AUC）、F1分数、阳性预测值、阴性预测值、布里尔分数、校准曲线和决策曲线分析来评估模型的性能。使用SHapley加性解释法来解释ML模型。
模型3的AUC优于模型1和模型2。在模型3中，AUC最高的前四种算法分别是梯度提升（0.81）、逻辑回归（0.81）、朴素贝叶斯（0.80）和随机森林（0.80）。对这四种算法进行进一步比较后发现，梯度提升略优于随机森林和逻辑回归，而朴素贝叶斯表现最差。
本研究使用生命体征、人口统计学、病史和主要症状开发了一种可解释的ML分诊模型，在预测ICU入住方面比传统模型更有效。可解释的ML有助于分诊过程中的临床决策。
