The logistic regression model based on age,gender,blood pressure,blood glucose and blood lipids can effectively predict white matter lesions of patients
LIN Yasen
MO Xianfu
CEN Huang
ZHOU Quan
Abstract：Objective To develop a logistic regression model for predicting the severity of white matter lesions of brain(Fazekas score)based on commonly used clinical indicators such as age,gender,blood pressure,blood glucose,and blood lipids,providing a reference for clinical screening and early intervention.Methods A retrospective analysis was conducted on 300 elderly patients who underwent MRI examinations in the Third Affiliated Hospital of Southern Medical University from January 2018 to October 2024.Clinical and laboratory data,including age,gender,systolic blood pressure,diastolic blood pressure,fasting blood glucose,low-density lipoprotein,triglycerides,and total cholesterol,were collected.The Fazekas score was ranked mainly using MRI T2-FLAIR sequence imaging data to assess the degree of white matter lesions in the brain,with mild white matter lesions defined as a Fazekas score of≤3 and severe white matter lesions defined as a Fazekas score of>3.Patients were randomly divided into a training set(n=240)and a test set(n=60).Clinical and laboratory data were used as independent variables,and the severity of white matter lesions as dependent variables to develop a logistic regression model.Model performance was assessed using the ROC curve and the AUC in both the training and test sets.Results Among the 300 patients,102(34.0%)had mild white matter lesions,while 198(66.0%)had severe lesions.Univariate analysis showed that patients in the severe group had significantly higher age,systolic blood pressure,diastolic blood pressure,and fasting blood glucose levels than those in the mild group(P<0.05),with an increased proportion of males.According to the Akaike Information Criterion,age,gender,systolic blood pressure,low-density lipoprotein,triglycerides,and total cholesterol were selected as independent variables for the logistic regression model.The model achieved an AUC of 0.778 in the training set,with a sensitivity of 0.763 and a specificity of 0.624.In the test set,the AUC was 0.860,with a sensitivity of 0.791 and a specificity of 0.688,demonstrating good predictive performance.Conclusion A prediction model based on age,gender,blood pressure,and blood glucose can effectively predict the severity of white matter lesions,with potential clinical value.
Keywords：white matter lesions of the brainFazekas ratingmagnetic resonance imagingpredictive modelageblood pressureblood lipids
Publication Date：2025-10-20
Online Publishing Date：2025-11-24(First online date of this platform, not the publication date of the document)
Pages：6( 1269-1274 )