Novel Deep Learning Architecture for Predicting Heart Disease using CNN

In the last few years, with increased population the most critical component of human life is healthcare. Compare to other deadly diseases, heart disease is one of the most lethal diseases, affecting the lives of millions of people worldwide. It is very important to detect heart disease must early so the loss of lives can be prevented. The availability of enormous amounts of data for medical diagnostics has aided in the development of complex learning-based models for automated early detection of cardiac problems. The classical machine learning approaches unable to generalize the new data sets which have not been seen in the training set. Therefore, the trained model has less accuracy in prediction stage. To minimize this issue, need to balance between training and testing datasets. This paper proposes a novel deep learning architecture using a 1D convolutional neural network for classification between healthy and non-healthy persons with balanced datasets to reduce the limitations of classical machine learning approach. Several clinical parameters are used for evaluating the risk contour in the patients which supports in early diagnosis. Various regularization methods are used to avoid overfitting in the proposed model. The proposed model achieves over 97% training accuracy and 96% test accuracy on the dataset. This is compared in detail with other machine learning algorithms using various performance parameters which proves the effectiveness of the proposed model.