Prediction and Early Detection of Heart Disease: A Hybrid Neural Network and SVM Approach

Heart disease is a leading cause of death worldwide, claiming approximately 17.9 million lives annually. Timely and accurate diagnosis is essential for reducing severe health outcomes, yet traditional methods are prone to delays and human error. This study presents a hybrid machine learning model that combines Neural Networks (NN) and Support Vector Machines (SVM) to improve the early detection and prediction of heart disease. The model leverages NN’s ability to capture complex, non-linear data patterns and SVM’s precision in classification to enhance diagnostic accuracy. Using a dataset of 303 patient records with 14 clinical features, the model underwent extensive preprocessing to optimize input data. The hybrid approach first applies NN for feature learning and then uses an SVM with an RBF kernel for classification. Testing revealed a 92% accuracy rate, outperforming standalone models and demonstrating strong performance in key metrics such as precision, recall, F1-score, and ROC-AUC. This system provides a reliable and efficient tool for healthcare professionals to facilitate early diagnosis and personalized treatment of heart disease. Future work will focus on integrating additional clinical and genetic data to improve further the model’s adaptability and predictive accuracy across diverse patient populations.