Prediction Heart Disease using Neural Network

Doctors typically cannot predict or prevent heart attacks in young people unless there are signs and symptoms that show up on a physical exam, echocardiogram, or electrocardiogram. A healthy young person does not always have physical signs of heart problems, so they go undetected until a heart attack occurs .We have proposed a neural network- based technique that can identify cardiac illness early on, helping to reduce the chance of death. The Cleveland Heart Disease dataset, which is publicly available, is used to test the suggested approach. Our model incorporates improvements such as the use of the Adam optimizer, which adapts the learning rate during training, and 10-fold cross-validation, a technique that enhances the model's generalizability and reduces the risk of over fitting. With these modifications, we have achieved an accuracy of 91.88%, demonstrating the effectiveness of our model in detecting heart disease and outperforming existing methods in the literature.