Coronary Artery Disease Prediction using Bayesian-Optimized Support Vector Machine with Feature Selection

Abdul Zahir Baratpur¹Hamed Vahdat-Nejad¹Emrah Arslan²Javad Hassannataj Joloudari¹Silvia Gaftandzhieva^3*

• ¹University of Birjand, Birjand, Iran
• ²KTO Karatay Universitesi, Konya, Türkiye
• ³Plovdivski universitet Paisij Hilendarski, Plovdiv, Bulgaria

Cardiovascular diseases, particularly Coronary Artery Disease (CAD), remain a leading cause of mortality worldwide. Invasive angiography, while accurate, is costly and risky. This study proposes a non-invasive, interpretable CAD prediction framework using the Z-Alizadeh Sani dataset. A hybrid decision tree–AdaBoost method selects 30 clinically relevant features. To prevent data leakage, SMOTE oversampling is applied exclusively within each training fold of a 10-fold cross-validation pipeline. The SVM is optimized using Bayesian hyperparameter tuning, compared against Sea Lion Optimization (SLOA) and grid search. Ablation studies and Wilcoxon signed-rank tests confirm the statistical superiority of the proposed SVM_Bayesian model. SHapley Additive exPlanations (SHAP) analysis reveals clinically meaningful feature contributions (e.g., Typical Chest Pain, Age, EF-TTE). 95% bootstrap confidence intervals and temporal generalization on an independent test set ensure robustness and no overfitting. The proposed model achieves 97.67% accuracy, 95.45% precision, 100.00% sensitivity, 97.67% F1-score, and 99.00% AUC, outperforming logistic regression (93.02% accuracy, 92.68% F1-score), random forest (95.45% accuracy, 93.33% F1-score), standard SVM (77.00% accuracy), and SLOA-optimized SVM (93.02% accuracy). Future work includes validation on external real-world datasets. This framework suggests a transparent, generalizable, and clinically actionable tool for CAD risk stratification, and is in keeping with the tenets of network physiology with its focus on interconnected cardiovascular features in predicting systemic disease.