急性冠脉综合征经皮冠状动脉介入治疗术后1年发生心力衰竭预测模型的构建及效果评价

Construction of prediction model of heart failure one year after percutaneous coronary intervention for acute coronary syndrome

目的探究急性冠脉综合征(ACS)经皮冠状动脉介入治疗(PCI)术后1年发生心力衰竭随机森林和logistic预测模型的构建及预测效果。方法将2021年6月至2022年6月于哈尔滨医科大学附属第一医院进行PCI手术的208例患者作为研究对象。根据术后1年是否发生心力衰竭分为心力衰竭组(n=31)和对照组(n=177),将筛选处理得到的影响因子应用随机森林和logistic回归算法分析ACS患者PCI术后1年发生心力衰竭的影响因素,采用受试者操作特征(ROC)曲线下面积(AUC)评价其预测效能。结果208例患者中心力衰竭患者31例(14.90%),采用logistic回归随机和森林算法构建的模型预测ACS患者PCI术后1年发生心力衰竭的AUC分别为0.902、0.967,预测效能均较好。结论基于ACS患者PCI术后1年发生心力衰竭的危险因素构建的预测模型具有较好的预测效能。

Objective To explore the construction and prediction effect of random forest and logistic regression model for heart failure one year after percutaneous coronary intervention(PCI)for acute coronary syndrome(ACS).Method 208 patients who underwent PCI in the First Affiliated Hospital of Harbin Medical University from June 2021 to June 2022 were taken as the research object.They were divided into heart failure group(n=31)and control group(n=177)according to whether heart failure occurred one year after PCI.The infl uencing factors of heart failure in ACS patients one year after PCI were analyzed by using random forest and logistic regression algorithm,and the predictive performance was evaluated by using the area under the operating characteristic(ROC)curve(AUC).Result Among the 208 patients,there were 31 patients with heart failure(14.90%).There was no signifi cant difference in AUC between logistic regression and random forest in predicting the occurrence of heart failure in ACS patients one year after PCI.Conclusion The prediction model based on the risk factors of heart failure in ACS patients one year after PCI can directly predict the probability of heart failure in ACS patients one year after PCI.