Objective This study aims to construct a prediction model for large artery atherosclerotic(LA A)ischemic stroke based on NOD-like receptor thermal protein domain associated protein 3(NLRP3)gene polymorphism and support vector machine(SVM)mod-el.Methods 84 LAA patients and 84healthy controls were selected to detect polymorphisms at rs10159239,rs10754558,rs12143966 and rs3806265 of NLRP3 gene,construct the SVM model,and the model performance was evaluated by receiver operating characteristic(ROC)curve.Results The genotype and allele frequency of NLRP3gene polymorphism sites were compared between the case group and the control group,and the difference was statistically significant(P＜0.05).The area under curve(AUC)of the ROC in the training and test sets is 0.900(95％CI:0.850-0.950)and 0.903(95％CI:0.807-0.999),respectively,with high prediction efficacy and no over-fit.The results of SHAP value analysis showed that:NLRP3 rs10159239genotype AG,rs10754558genotype CG and GG,rs12143966genotype GG and rs3806265genotype CT are the risk factors for disease development.Conclusions NLRP3gene polymor-phism is closely related to the risk of LA ischemic stroke.SVM model based on NLRP3 gene polymorphism can be used to predict the risk factors of LAA ischemic stroke,providing a new idea for early clinical diagnosis and precise prevention.

Objective This study aims to construct a prediction model for large artery atherosclerotic(LA A)ischemic stroke based on NOD-like receptor thermal protein domain associated protein 3(NLRP3)gene polymorphism and support vector machine(SVM)mod-el.Methods 84 LAA patients and 84healthy controls were selected to detect polymorphisms at rs10159239,rs10754558,rs12143966 and rs3806265 of NLRP3 gene,construct the SVM model,and the model performance was evaluated by receiver operating characteristic(ROC)curve.Results The genotype and allele frequency of NLRP3gene polymorphism sites were compared between the case group and the control group,and the difference was statistically significant(P＜0.05).The area under curve(AUC)of the ROC in the training and test sets is 0.900(95％CI:0.850-0.950)and 0.903(95％CI:0.807-0.999),respectively,with high prediction efficacy and no over-fit.The results of SHAP value analysis showed that:NLRP3 rs10159239genotype AG,rs10754558genotype CG and GG,rs12143966genotype GG and rs3806265genotype CT are the risk factors for disease development.Conclusions NLRP3gene polymor-phism is closely related to the risk of LA ischemic stroke.SVM model based on NLRP3 gene polymorphism can be used to predict the risk factors of LAA ischemic stroke,providing a new idea for early clinical diagnosis and precise prevention.