Hybrid Model for Improved Heart Disease Prediction

Abstract

Cardiovascular disease, which encompasses various conditions affecting the heart and blood vessels, is a significant global health concern and a primary 
cause of mortality on a global scale. These ailments have a profound impact on heart function, blood circulation, and overall well-being. This investigation 
introduces a novel hybrid model that effectively combines the strengths of Decision Tree (DT), Logistic Regression (LR), and Artificial Neural Network 
(ANN) algorithms, thereby significantly augmenting the accuracy of heart disease prediction. The model demonstrates exceptional performance, boasting an impressive accuracy rate of 88%, which surpasses the individual accuracies of DT at 99%, LR at 80%, and ANN at 86%. Furthermore, the hybrid approach excels in precision, recall, and F1-score metrics, thereby substantiating its reliability and robustness as a predictive tool for heart disease. This research underscores the advantages of incorporating multiple algorithms in order to create a more efficient predictive model for cardiovascular health diagnostics.