中国实用神经疾病杂志2025，Vol.28Issue(12)：1460-1465,6.DOI:10.12083/SYSJ.250632
基于随机森林算法预测急性脑梗死溶栓后血管再闭塞的临床研究
Clinical study of predicting vascular reocclusion after thrombolysis for acute cerebral infarction based on random forest algorithm
摘要
Abstract
Objective To investigate the influencing factors of vascular reocclusion after thrombolysis in acute cerebral infarction(ACI),and to construct a random forest algorithm model.Methods A total of 252 ACI patients admitted to the Beijing Friendship Hospital Affiliated to Capital Medical University from January 2022 to December 2024 were selected as study subjects.They were categorized into the occurrence group and the non-occurrence group based on the occurrence of vessel reocclusion after thrombolysis.The general information of the two groups was compared,and the influencing factors for vessel reocclusion after ACI thrombolysis were analyzed using Logistic regression equations.A Logistic regression model was constructed,and a random forest algorithm model for vessel reocclusion after ACI thrombolysis was established using R software.Receiver operating characteristic(ROC)curves and calibration curves were plotted to analyze the predictive performance of the two models.Results The NIHSS score,SIRI,miR-127,miR-320,and hyperglycemia proportion at admission in the occurrence group were all higher than those in the non-occurrence group(P<0.05).The Logistic regression equation showed that NIHSS score,SIRI,miR-127,miR-320,and hyperglycemia at admission were influencing factors for vascular reocclusion after thrombolysis in ACI patients(P<0.05).The random forest model indicated that the optimal model could be obtained when ntree=300,and the importance of the influencing factors for vascular reocclusion after thrombolysis in ACI patients was ranked as NIHSS score,miR-320,SIRI,miR-127,and hyperglycemia.The AUC of the random forest algorithm model for predicting vascular reocclusion after thrombolysis in ACI patients was higher than that of the Logistic regression equation model,and the calibration curve fluctuated around the diagonal.Conclusion The ACI post-thrombolysis vessel reocclusion early warning model developed based on the random forest algorithm has good predictive performance,which is helpful for formulating relevant prevention and control measures early and reducing the risk of vessel reocclusion.关键词
急性脑梗死/溶栓/血管再闭塞/随机森林算法/Logistic回归模型Key words
Acute cerebral infarction/Thrombolysis/Vascular reocclusion/Random forest algorithm/Logis-tic regression model分类
医药卫生引用本文复制引用
佟珊,刘潮,刘晓辰,张志月..基于随机森林算法预测急性脑梗死溶栓后血管再闭塞的临床研究[J].中国实用神经疾病杂志,2025,28(12):1460-1465,6.基金项目
河北省医学科学研究重点课题计划项目(编号:20231794) （编号:20231794）
