Heart Disease Prediction Using Deep Learning and Ensemble Machine Learning Models

Abstract

Heart disease remains one of the leading causes of mortality worldwide, necessitating the development of accurate and reliable prediction systems. This study presents a comparative analysis of several ensemble machine-learning algorithms and a deep neural network (DNN) model for the prediction of heart disease using clinical data. The models evaluated include Random Forest, AdaBoost, Gradient Boosting, XGBoost, LightGBM, and a deep-learning model built using TensorFlow. The dataset comprises patient records with attributes such as age, cholesterol level, resting blood pressure, and exercise-induced angina. Performance was evaluated using standard metrics including accuracy, precision, recall, and F1- score. Among all the models, the Deep Neural Network achieved the highest performance with an accuracy of 98%, outperforming the best ensemble-based model, XGBoost, which achieved 97%. The results demonstrate that deep learning can offer superior predictive capability for cardiovascular risk detection, making it a viable choice for real-world clinical decision-support systems. The system is implemented in Python with a Django web framework and a relational database, integrating data preprocessing, model training, evaluation, and a web interface for prediction. By combining traditional ensemble techniques with a deep neural network, the proposed system enhances the reliability of medical prediction and supports early detection and prevention of heart disease