Survey on Machine Learning Classifiers for Detecting Rheumatic Heart Disease (RHD) using the UCI Debrecen Dataset

Rheumatic Heart Disease (RHD) remains a major cause of preventable cardiovascular morbidity and mortality, particularly among children and young adults in lowand middle-income countries. Early detection of valvular damage is essential for preventing disease progression; however, conventional echocardiographic diagnosis is resource-intensive and highly reliant on expert interpretation.
Recent advances in artificial intelligence (AI) and machine learning (ML) offer promising opportunities to automate RHD screening and support equitable access to care. Clinical
Decision Support Systems (CDSS) can assist clinicians in timely and accurate diagnosis, though selecting an effective ML classifier remains a key challenge. This study evaluates five ML algorithms—Logistic Regression, Gradient Boosting, Random Forest, K-Nearest Neighbors (KNN), and Support Vector Machine—using the UCI Debrecen dataset. Models were assessed using precision, accuracy, F1-score, and recall. Random Forest achieved the best performance, with an F1-score of 72.17% and accuracy of 72.29%, indicating strong overall classification capability. KNN showed the poorest performance, with 60.17% accuracy and the lowest recall (54.47%), highlighting its limited ability to capture complex data patterns.