Detection of Heart Sound Abnormality using Artificial Intelligence & Machine Learning

Abstract

Phonocardiogram signals – or PCG signals – are the recordings of sounds and murmurs created by the heart. They are used to detect heart abnormalities in a clinical environment. In this paper, we try to automate the detection process. We look at heart signal recordings and analyze them using advanced machine learning algorithms and models. These algorithms are capable of identifying subtle patterns and anomalies in the heart signal data that may be indicative of underlying cardiac conditions. By automating the process, we can process large volumes of data in a fraction of the time it would traditionally take to do the same. This not only increases the efficiency of our analysis but also allows us to deliver results in real-time, which could be critical in urgent care situations. Also, incorporating machine learning models into the detection mechanism allows the machine to continue developing by looking at more samples and become better at predicting abnormalities. It will also allow the model to incorporate any new heart abnormalities in the future without much difficulty