Cardiovascular Disease Prediction Using Deep Learning Models

One of the most prevalent and significant diseases affecting people's health is cardiovascular disease (CVD). In early diagnosis, cardiovascular may be less severe and this may decrease the death rates. Distinguishing risk factors utilizing deep learning models is a hopeful methodology. Our research is based on the “Efficient Prediction of Cardiovascular Disease Using Machine Learning Algorithms With Relief and LASSO Feature Selection Techniques” gives the accuracy of 99.05% and we have extended this research by applying deep learning models, ANN, and CNN for prediction of heart disease from their health measures. L2 regularization technique is applied on these models to handle overfitting problem and the same dataset is applied for getting more accuracy in the result so that the physician diagnoses the ailment with confidence. To generate reliable data for the training model, we effectively collected data, processed data, and transformed data. For the evaluation of the proposed method, the metrics reveal accuracy, sensitivity, precision, recall, and F1-Score. The proposed method, ANN, gives 98% accuracy, and the CNN method gives 95% accuracy with L2 regularization. Hence, the ANN model gives the best accuracy result.