Predicting Cardiac Diseases Through Big Data Analytics Using Artificial Neural Networks in Healthcare

Cardiovascular diseases (CVDs) remain one of the leading causes of mortality worldwide, necessitating efficient early prediction systems to support timely clinical intervention. This study presents a robust cardiac disease prediction framework leveraging Big Data Analytics and Artificial Neural Networks (ANNs), trained and validated using the Cleveland Heart Disease dataset. The methodology includes comprehensive data preprocessing involving imputation, normalization, and encoding of clinical and lifestyle features such as age, blood pressure, cholesterol, physical activity, and smoking habits. To enhance prediction accuracy and reduce redundancy, Recursive Feature Elimination (RFE) and Principal Component Analysis (PCA) were employed for feature selection and dimensionality reduction. The optimized ANN model, structured with multiple hidden layers and a sigmoid-activated output, was trained to perform binary classification of heart disease risk. The model was evaluated using five-fold cross-validation and key performance metrics, achieving an accuracy of 89.3%, precision of 90.2%, recall of 88.5%, F1-score of 89.3%, and an AUC of 0.92. These results outperform several traditional classifiers, including SVM, Decision Trees, Logistic Regression, and even hybrid models, demonstrating the ANN's superior ability to model complex, nonlinear relationships in healthcare data. The proposed approach shows strong potential for real-world clinical deployment, offering an effective, scalable, and interpretable tool for early cardiac risk detection.