Heart murmur detection/classification using Cochlea-like pre-processing.

Abstract

Accurate detection and classification of pathological heart murmurs by auscultation has been a challenge for physicians for a long time. Many research efforts have been made to apply artificial intelligence (AI) for rigorous detection/classification of heart murmurs but reported success rates have been low. All of the current AI techniques rely on converting the heart sounds to electrical signals and processing those signals via electronic circuitry of AI for murmur detection and classification. In this research, we have used a novel approach to pre-process the electrical heart sound signals by altering the electrical signal in a similar way as is done by human cochlea before they go to AI for murmur detection/classification. Cochlea-like pre-processing changes the spectral contents of the heart sound signal to enhance the murmur information which can then be detected and classified more accurately by AI circuitry. We have designed a heart murmur detection/classification system based upon this approach and have tested this system using simulated heart sounds of various murmur types. Our test results show that this approach significantly improves heart murmur detection/classification accuracy.