An Ensemble-based ML Model to Predict Cardiac Disease

In contemporary times cardiovascular disease has emerged as a predominant health concern. It underscores a critical necessity for early and precise predictions to facilitate effective prevention and intervention strategies. This investigation is centered on a meticulous comparative analysis of heart disease prediction which employs advanced machine learning techniques. Three distinct machine learning models such as Random Forest, Light GBM and Gradient Boosting are deployed to analyze and predict heart disease based on an available dataset. The study commences with comprehensive data preprocessing, incorporating sophisticated sampling techniques. Subsequently, the dataset is stratified into training and testing sets, and the three machine learning models are systematically implemented. Evaluation metrics, including accuracy, precision, recall, and F1- score, are employed to assess the predictive capabilities of the models. Three combinations are selected for ensemble modeling based on their better performance for the used dataset. Notably, the ensemble model of Random Forest and XGBoost attained a peak accuracy of 99%. The study highlights the use of machine learning, particularly ensemble model in enhancing diagnostic accuracy for heart disease prediction, holding the potential to significantly contribute to early detection and timely intervention, thereby alleviating healthcare burdens related to cardiovascular health.