Classification of Heart Sounds using Machine Learning Algorithm

The accurate diagnosis of heart abnormalities is crucial for the early detection and treatment of cardiovascular diseases. Conventional auscultation methods relying on human experts are subjective and time-consuming. To address this challenge, this study explores the application of machine learning algorithms for the automated classification of heart sounds. The research aims to develop a low-cost, efficient solution that can assist healthcare professionals in identifying potential heart ailments through the analysis of heart sound recordings. This machine learning algorithm uses the frequency, amplitude, and temporal characteristics of heart sounds to discriminate between normal and abnormal heart conditions, including murmurs and extrasystoles. And it follows by proposing a novel machine learning algorithm based on Recurrent Neural Network (RNN) and Convolutional Neural Network which demonstrates promising results in classifying different types of heart sounds. Long-Short-Term Memory (LSTM) and Convolutional Neural Network (CNN) models were employed for this purpose, achieving high accuracy in differentiating normal, murmur, and artifact sounds. The findings of this study contribute to the development of a mobile solution for the early detection of cardiovascular diseases, which can potentially improve patient outcomes and reduce healthcare costs.