<![CDATA[Coronary heart disease (CHD) remains one of the leading causes of death worldwide, highlighting the urgent need for accurate early detection. This study aims to compare the performance of various machine learning models—including Decision Tree, K-Nearest Neighbor (KNN), Logistic Regression, Random Forest, XGBoost, and Stacking Ensemble—in predicting CHD using the UCI Heart Disease Dataset. The models were evaluated using five metrics: accuracy, precision, recall, F1-score, and AUC. The results indicate that Stacking and Logistic Regression achieved the highest AUC scores (0.80), while XGBoost obtained the best F1-score (0.40). Simpler models such as Decision Tree and KNN showed relatively lower performance. In addition, feature importance analysis using permutation methods revealed that features like number of major vessels (ca), thalassemia (thal), ST depression (oldpeak), and age play a critical role in prediction accuracy. These findings demonstrate that ensemble learning approaches, especially Stacking and XGBoost, can effectively improve diagnostic performance and offer strong potential for clinical decision support systems (CDSS) in the early detection of coronary heart disease.]]>
Copyrights © 2025
