Cardio Care: A Predictive Model for Heart Disease Detection

Heart Disease is still an epidemic global health concern, so trustworthy prediction models are vital for risk assessment and early detection. In this research, we propose a comprehensive predictive framework leveraging ensemble learning techniques alongside traditional Machine Learning (ML) models to increase forecasting of cardio disease accuracy. Utilizing a dataset sourced from Cardiovascular_Disease_Dataset, demographic, clinical, and lifestyle factors were incorporated as features. The performance of Naive Bayes (NB), AdaBoost, Support Vector Classifier (SVC) and Logistic Regression (LR) both individually and within ensemble frameworks, was evaluated using accuracy of 98% and AUROC (Area-Under the Receiver Operating Characteristic Curve) of 99.845% as a primary metric. Our results demonstrate how effective ensemble learning strategies are in obtaining higher accuracy, precision-recall and along with AUROC curve for heart disease prediction compared to individual classifiers. This study underscores the importance of ensemble methods in improving predictive performance and contributing to enhanced clinical decision-making in cardiovascular health management.