Heart Murmur Detection using Artificial Neural Network

Abstract

Accurate detection and classification of pathological heart murmurs by auscultation has been a challenge for physicians for a long time. In this research, we have used a AI techniques rely on converting the heart sounds to electrical signals and processing those signals by matlab tool box for murmur detection and classification. We have designed a heart murmur detection system based upon this approach and have tested this system using simulated heart sounds of various murmur types. Our test results show that the used of multilayer perceptron ANNs improve heart murmur detection accuracy up to (97%).