急性穿支动脉脑梗死患者血清CysC hs-CRP联合随机森林预测END风险OA

Objective To investigate the value of constructing a random forest model based on serum cystatin C(CysC),high-sensitivity C-reactive protein(hs-CRP),and other indicators for predicting the risk of early neurological deterioration(END)in patients with acute perforating artery cerebral infarction(APACI).Methods A total of 189 APACI patients at Jiaozuo Second People's Hospital from August 2023 to August 2025 were retrospectively enrolled and divided into two groups according to the occurrence of END.Clinical data were compared between the two groups.Binary Logistic regression was used to analyze independent risk factors for END,and these factors were used as feature variables to construct a random forest model.Model robustness was evaluated by 10-fold cross-validation,and predictive performance was assessed using the receiver operating characteristic curve.Results Levels of low-density lipoprotein cholesterol(LDL-C),CysC,hs-CRP,and uric acid(UA)were significantly higher in the END group than those in the non-END group,and all were identified as independent risk factors for END.The random forest model ranked the feature importance in the order of CysC,hs-CRP,UA,and LDL-C.The model achieved the area under curve of 0.866,the sensitivity of 0.733,and the specificity of 0.833 for predicting END.Conclusion The random forest model based on serum CysC,hs-CRP,LDL-C,and UA levels at admission can effectively predict the risk of END in APACI patients.