Establishment and validation of a risk prediction model for acute ischemic stroke in a carotid artery plaque population.

Background: The accurate and timely prediction of acute ischemic stroke (AIS) is essential to optimize treatment strategies. However, there are many risk factors closely associated with AIS, and the research methods are varied and incomplete. This study developed a prediction model based on ultrasonic imaging and clinical parameters, and evaluated its clinical application value as a quantitative tool for predicting the risk of AIS and guiding clinical prevention and treatment. Methods: A cross-sectional study was conducted to assess the risk factors associated with AIS. Univariate, multivariate, and least absolute shrinkage and selection operator (LASSO) regression analyses were conducted to screen variables, identify relevant independent risk factors, and construct a nomogram prediction model. In addition, receiver operating characteristic (ROC) curves were used to assess model accuracy, C statistics were used to assess model differentiability, and calibration curves were used to assess clinical consistency. Results: A total of 244 patients were included in the study. Significant variables (P<0.05) identified by the univariate and LASSO regression analyses were included in the multivariate analysis, and five variables, including the plaque area, neovascularization density, white blood cell (WBC) count, waist ratio, and a history of alcohol consumption, were included the prediction model. A column-line prediction model was successfully constructed. The ROC curves of the training set and the test set were 0.738 [95% confidence interval (CI): 0.582-0.884] and 0.733 (95% CI: 0.669-0.807), respectively, indicating good accuracy. The C statistic was >0.7, indicating that the ability of the model to identify patients with AIS was relatively good. The calibration curve showed good consistency with the ideal curve, and the degree of fit was good. The P values of the Hosmer and Lemeshow tests of the model were both >0.05, indicating that the calibration degree of the prediction model was good. The probability of AIS predicted by the prediction model was in good agreement with the actual probability. The decision curve analysis (DCA) showed that the model had good Clinical practicality. Conclusions: The plaque area, the density of new blood vessels, WBC count, girth ratio, and a history of alcohol consumption were identified as independent risk factors and included in a nomogram prediction model for AIS.