Early Coronary Heart Disease Deciphered via Support Vector Machines: Insights from Experiments

Contemporary global health is increasingly shaped by Coronary Heart Disease (CHD), a swiftly growing concern. Precisely predicting CHD remains crucial for optimal patient care. This study analyzes various classifiers to sort through a complex dataset related to CHD, meticulously curated and computationally characterized with 919 patient cases. Narrowing the focus to a subset of 12 attributes, each annotated, a support vector machine (SVM) classifier is applied across different models—Linear, Polynomial, Radial Basis Function (RBF), and Sigmoid. Results show the SVM Linear model achieving 87.7% accuracy, while Polynomial, RBF, and Sigmoid models achieve 81.6%, 83.3%, and 87.8% respectively. Metrics like Area Under the Curve (AUC), sensitivity, and specificity add quantitative depth. Feature reduction's potential impact on precision and classifier efficiency is acknowledged, emphasizing ML's crucial role in refining diagnostic precision for cardiac conditions. The study offers a comprehensive view of cardiovascular ailments across age groups, empowering clinicians with data-driven medical insights via advanced machine learning.