HSCAD:Heart Sound Classification for Accurate Diagnosis using Machine Learning and MATLAB

With the rise in technology AI (Artificial Intelligence) is being used in healthcare at a wider level. Also, for health care automation deep learning and medical image processing is being proven as a boon for capturing and tracking accurate patient record and disease prediction. Heart sounds play a significant role in screening for heart disease. Because the signal-to-noise ratio (SNR) is low; it is a problem that separates experts and takes time to estimate correct heart waves. Also, ECG verification is being done prominently for every kind of disease and abnormality, which is why monitoring accurate ECG and heart rate is very essential in proper diagnosis. Distinguishing the purpose of the heart is therefore essential. Here in the study, we combined conventional feature engineering methods with deep learning to automatically distinguish between normal and abnormal heart sounds. We proposed a one-dimensional convolution neural network (CNN) model that divides the ECG into abnormal heart sounds. We have used multilayer perceptron (MLP) for the classification of heart waves. Differentiation of heart sounds is implemented using the MATLAB framework.