护理学报2025，Vol.32Issue(9)：64-68,5.DOI:10.16460/j.issn2097-6569.2025.09.064
基于机器学习算法的非体外循环冠状动脉旁路移植术患者术中低体温风险预测模型的构建
Construction of risk prediction model for intraoperative hypothermia in patients undergoing off-pump coronary artery bypass grafting based on machine learning algorithms
张梦冉 1李妍 1张梦晗 1张増梅 2张娟1
作者信息
- 1. 郑州大学第一附属医院麻醉科,河南郑州 450052
- 2. 郑州大学第一附属医院手术部,河南郑州 450052
- 折叠
摘要
Abstract
Objective To construct a risk prediction model of intraoperative hypothermia in patients undergoing off-pump coronary artery bypass grafting(OPCABG)using four kinds of machine learning algorithms including back propagation neural network(BPNN),support vector machine(SVM),random forest(RF)and logistic regression(LR),and to provide reference for evaluating and preventing intraoperative hypothermia.Methods A total of 1,065 patients with OPCABG in one hospital in Henan Province from January 2021 to July 2024 were selected as research objects.Least absolute shrinkage and selection operator(LASSO)and multivariate logistic regression were used to screen the predictors.Four kinds of machine learning algorithms were used to construct the risk prediction model of intraoperative hypothermia,and the receiver operating characteristic curve area(AUC)was used to evaluate the performance of the model,and SHAP for explanatory analysis.Results The AUC values of the BPNN,SVM,RF,and LR models were 0.821,0.822,0.978,and 0.730,respectively,with the model constructed by RF performing the best.Conclusion The RF-based risk prediction model of intraoperative hypothermia in OPCABG patients has the best performance,which is helpful for early evaluation and prevention of intraoperative hypothermia.关键词
机器学习/随机森林/非体外循环冠状动脉旁路移植术/非计划术中低体温/预测模型Key words
machine learning/random forest/off-pump coronary artery bypass graft/inadventent intraoperative hypothermia/pre-diction model分类
医药卫生引用本文复制引用
张梦冉,李妍,张梦晗,张増梅,张娟..基于机器学习算法的非体外循环冠状动脉旁路移植术患者术中低体温风险预测模型的构建[J].护理学报,2025,32(9):64-68,5.
