基于机器学习建立腹部手术术后脓毒症患者死亡风险预测模型  

Prediction of postoperative sepsis mortality risk based on machine learning in patients undergoing abdominal surgery

作　　者：舒欣 李昊洋 李雨捷 宋艾璘 胡小艳 陈芋文 张炬[3] 易斌 SHU Xin;LI Haoyang;LI Yujie;SONG Ailin;HU Xiaoyan;CHEN Yuwen;ZHANG Ju;YI Bin(Department of Anesthesiology,First Affiliated Hospital,Army Medical University(Third Military Medical University),Chongqing,400038;Regiment Five,Basical Medicine College,Army Medical University(Third Military Medical University),Chongqing,400038;Chongqing Institute of Green and Intelligent Technology,Chinese Academy of Sciences,Chongqing,400714,China)

机构地区：[1]陆军军医大学(第三军医大学)第一附属医院麻醉科,重庆400038 [2]陆军军医大学(第三军医大学)基础医学院学员五大队,重庆400038 [3]中国科学院重庆绿色智能技术研究院,重庆400714

出　　处：《陆军军医大学学报》2023年第8期732-738,共7页Journal of Army Medical University

基　　金：国家重点研发计划课题(2018YFC0116702);重庆英才计划“包干制”项目(CQYC202103080)。

摘　　要：目的探讨机器学习算法构建腹部手术术后脓毒症患者死亡风险预测模型的可行性。方法采用病例-对照研究设计方案,从公共重症监护医学信息数据库(Medical Information Mart for Intensive CareⅣ,MIMIC-Ⅳv1.0)中筛选出行腹部手术后发生脓毒症的患者,研究终点事件定义为患者入院后90 d内死亡。根据死亡与否将数据集随机拆分为训练数据集(70%)与测试数据集(30%),在训练数据集上基于Logistic回归(logistic regression,LR)、梯度提升树(gradient boosting decision tree,GBDT)、随机森林(random forest,RF)、支持向量机(support vector machine,SVM)和自适应提升(adaptive boosting,AdaBoost)算法构建预测死亡风险模型;在测试数据集上通过受试者工作曲线(receiver operating characteristic curve,ROC)和曲线下面积(area under the ROC curve,AUC)、敏感性、特异性、阳性预测值、阴性预测值、F1分数和准确率来评估模型效能。结果最终986例患者纳入本研究,其中251例(25.5%)患者入院后90 d内死亡,LR、GBDT、RF、SVM及AdaBoost模型的AUC依次为0.852、0.903、0.921、0.940和0.906,其中SVM的AUC最高,预测性能更好,而LR模型效能最差。结论基于GBDT、RF、SVM及AdaBoost这4种算法建立的腹部手术术后脓毒症死亡率预测模型的效能优于传统的LR模型,可能有助于临床决策,改善不良结局。ObjectiveTo explore the feasibility of constructing prediction models of postoperative sepsis mortality risk based on machine learning in patients undergoing abdominal surgery.MethodsA case-control trial was designed and conducted on the patients diagnosed with sepsis after abdominal surgery from Medical Information Mart for Intensive Care Ⅳ(MIMIC-Ⅳ)database,and 90-day mortality was defined as the primary endpoint event after hospitalization.The dataset was ramdomly split into training(70%)and test(30%)datasets according to wether diagnosed with postopertive sepsis or not.On the training dataset,logistic regression(LR),gradient boosting decision tree(GBDT),random forest(RF),support vector machine(SVM)and adaptive boosting(AdaBoost)were used to develop the prediction model for death.The area under the receiver operating characteristic curve(AUC),sensitivity,specificity,positive predictive value,negative predictive value,accuracy and F1 score were used for model evaluation on the test dataset.ResultsA total of 986 patients were finally analyzed,of whom 251 patients(25.5%)died within 90 d after hospitalization.The AUC values of LR,GBDT,RF,SVM and AdaBoost prediction models were 0.852,0.903,0.921,0.940 and 0.906,respectively.The model based on SVM yielded the best AUC value,higher differentiation and better prediction performance,while LR performed the worst among them.ConclusionThe performances of the prediction model of postoperative sepsis mortality based on GBDTT,RF,SVM and AdaBoost are all better than that of traditional LR model,which may help to assist clinical decision making and improve adverse outcomes.

关 键 词：术后脓毒症 机器学习 腹部手术 预测模型 死亡风险 

分 类 号：R319[医药卫生—基础医学] R619.3R656.07