Heart Disease Prediction using CNN: Performance Comparison and Analysis

This study evaluates and compares various analysis techniques for convolutional neural networks (CNNs) in predicting cardiac illnesses. Several CNN architectures were developed, and their performance was thoroughly examined. The authors have also compared the outcomes of CNN-based strategies with those of other deep learning (DL) methods and traditional machine learning (ML) techniques. Our investigation highlighted the advantages and disadvantages of using CNNs to process and classify cardiovascular data. The findings indicated that CNNs performed better in terms of accuracy, especially when hyperparameter optimization techniques were combined with sophisticated preprocessing methods. However, we have also noted challenges such as overfitting and high computational demands. The authors have used the publicly available UCI heart disease dataset to assess the proposed model, which includes 14 features and 1,025 patients. The model was evaluated using various metrics, resulting in precision, accuracy, F1-score, and recall, which achieved the highest accuracy of 95.12%. The conclusions drawn from this research contribute to the development of diagnostic tools for the early detection of cardiac disease, ultimately aiming to improve patient outcomes and inform treatment strategies.