Predicting Cardiovascular Disease Risk and Outcomes Using a Graph Convolutional Network Algorithm

Cardiovascular disease continues to be a primary cause of death globally, rendering precise and prompt prognosis essential for successful healthcare measures. Deep learning (DL) methodologies have arisen as potent instruments for forecasting cardiovascular illness, facilitating early detection and intervention. This research introduces an innovative way for predicting cardiac illness through the application of advanced DL algorithms. The suggested method tackles class imbalance by employing the Synthetic Minority Over-sampling Technique (SMOTE), which equalizes the dataset by creating synthetic samples for the minority class. Using Recursive Feature Elimination (RFE) improves model performance by reducing noise and enhancing generalization through the identification of relevant features. A Graph Convolutional Network (GCN) is employed for classification, allowing the model to adeptly capture intricate interactions among features within a graph structure. A comparison examination of existing techniques, such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and Deep Belief Networks (DBNs), reveals that the suggested method attains a remarkable accuracy of 99.60%, underscoring its effectiveness in predicting heart disease. Also achieves 97.57% of specificity, 98.20% of precision, 98.59% of sensitivity and 99.60% of F1 score.