A Study on Analyzing Risk Factors for Carotid Plaque Instability Using Machine Learning Models

Objective: The objective of this study was to develop and evaluate the effectiveness of machine learning (ML) models in predicting carotid plaque instability. Methods: A retrospective analysis was conducted on data from 449 patients with carotid plaques treated at The Affiliated Hospital of Qingdao University between July 2022 and November 2024. Patients were randomly divided into a training set and a testing set at a 7:3 ratio. Five ML algorithms were utilized to establish prediction models. The predictive performance of each model was assessed using the area under the receiver operating characteristic curve (AUC), accuracy, sensitivity, and specificity on the testing set. Results: Among the five ML algorithms, the light gradient boosting machine (LightGBM) demonstrated the best performance with an AUC of 0.921 (95% confidence interval [CI]: 0.887–0.956) and an accuracy of 0.837 in the training set, and an AUC of 0.864 (95% CI: 0.784–0.944) and an accuracy of 0.811 in the testing set. The feature importance results showed that C-reactive protein (CRP), leukocytes, high-density lipoprotein (HDL), and platelets were the four most significant contributors to carotid plaque instability. Conclusion: We developed and validated the ML models for predicting carotid plaque instability, and identified that LightGBM model with 9 relevant factors had the best performance and high levels of clinical applicability.