Machine Learning Models for Predicting Heart Disease from Patient Data

Heart disease is among the foremost causes of mortality and morbidity worldwide, claiming an estimated 18 million lives annually. With the growing volume of healthcare data generated from clinical examinations, laboratory reports, and electronic health records, machine learning (ML) has emerged as a transformative approach for early disease prediction and risk stratification. This research investigates six supervised ML algorithms—Logistic Regression, Decision Tree, Random Forest, Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Artificial Neural Network (ANN)—applied to the Cleveland Heart Disease dataset. A comprehensive pipeline encompassing data preprocessing, model optimization, and cross-validation was implemented. Performance was measured using multiple metrics, including accuracy, precision, recall, F1-score, and ROC-AUC. Results indicate that ensemble and deep learning approaches substantially outperform linear models, with XGBoost achieving the highest overall predictive power (accuracy = 90.2%, ROC-AUC = 0.94). Beyond raw performance, the study emphasizes the ethical imperatives of interpretability, fairness, and clinical trust in deploying ML systems in healthcare. Findings support the integration of ML-based tools into clinical practice for early cardiovascular diagnosis and patient-specific risk management.