Early detection of heart murmur using advanced stethoscope: a machine learning and IoT prospective

Heart murmurs, indicative of underlying cardiac abnormalities, have traditionally been assessed through auscultation with a stethoscope. However, the subjective nature of this method can lead to variability in diagnosis, and the need for a machine-automated system becomes paramount. The proposed research aimed at developing an advanced stethoscope capable of detecting both heart sounds, specifically focusing on automating the detection of heart murmurs using the Internet of Things. The recorded heart sounds are then transformed into various parameters and stored in a database for further analysis. Furthermore, data preprocessing techniques are applied to handle missing data and encode the data. The study found that implementation of the Decision Tree Regression model yielded exceptional accuracy in screening for cardiac murmurs, showcasing its potential as a front-line clinical support tool. Illustrated graphs of heart sounds further enhance diagnostic capabilities, allowing for easy comparison with existing reference graphs.