Designing a machine learning model for predicting cardiovascular events using the triglyceride-glucose index: a cohort study - PubMed
3 hours ago
- #TyG Index
- #Machine Learning
- #Cardiovascular Disease
- Cardiovascular diseases (CVD) are the leading cause of death in developing countries.
- Early detection of high-risk CVD patients could reduce mortality.
- The study explores the triglyceride-glucose (TyG) index's effectiveness in predicting CVD events using machine learning.
- Data from the MASHAD cohort was used, with patients monitored for over ten years.
- Eleven machine learning models were evaluated, including MLP and AdaBoost classifier.
- CVD event prevalence was 10.9%, with an average age of 48.08 years and 60% female participants.
- The mean TyG index was 8.59 ± 0.66.
- MLP and AdaBoost models showed the highest predictive accuracy (ROC-AUC scores of 0.77 and 0.766).
- The TyG index was the fourth most significant predictor in MLP and AdaBoost models.
- Incorporating the TyG index could enhance CVD risk prediction accuracy, especially in developing countries.