Establishment Of Nomogram Model For Individualized Prediction Of Ischemic Stroke Risk

Objective:The clinical data of patients with ischemic stroke were analyzed retrospectively,the risk factors of previous ischemic stroke were analyzed and verified,the influencing factors that can predict the risk of ischemic stroke were screened out,and an individualized nomogram model for predicting the risk of ischemic stroke was designed and established.Methods:192 inpatients and physical examiners in neurology department of our hospital were divided into stroke group(133 cases)and non-stroke group(59 cases),including and collecting relevant medical history.The basic demographic data included age,gender,height,weight,whether suffering from hypertension,whether suffering from diabetes,etc.Relevant clinical laboratory data,such as homocysteine,high density lipoprotein cholesterol,serum amyloid A,apolipoprotein AI,triglyceride,total cholesterol,etc.Single factor analysis and Lasso regression were used to screen out the relevant risk factors of ischemic stroke from the relevant clinical indicators,and the above variables with statistical significance were incorporated into the multivariate logistic regression analysis model.The nomogram model for predicting the risk of ischemic stroke was established by R software,verified by Bootstrap method,and the ROC curve was used to explore the efficiency of nomogram model for predicting ischemic stroke.Results:Through Lasso regression and multivariate regression analysis,Age,diabetes mellitus(DM),high density lipoprotein cholesterol(HDL-C),homocysteine(Hcy)and serum amyloid A/ apolipoprotein AI(SAA/Apo A-I)were jointly selected as risk factors for ischemic stroke.The concordant C index of nomogram model for predicting ischemic stroke risk is 0.983,with good discrimination.Nomogram Forecast Model for Decision Curve Analysis Has High Net Benefit.Conclusion:This study was based on five risk factors for ischemic stroke,including age,diabetes,high-density lipoprotein cholesterol,homocysteine,serum amyloid A / apolipoprotein AI,etc.The nomogram model of the risk prediction of ischemic stroke has better discrimination and accuracy,and has higher clinical value.