Identification of aortic stenosis and mitral regurgitation by heart sound analysis

Abstract

Heart sound recording and analysis have the potential to provide objective results to assist clinicians with auscultation. A method is proposed for distinguishing the systolic murmurs arising from aortic stenosis and mitral regurgitation. The problems encountered include accounting for background noise due to breathing and other sources, and dealing with the beat to beat variation in the loudness of the murmur. The approach taken is to establish that a murmur is present and detectable within the noise, and then to analyse the recordings using a frequency spectrum analysis and a bank of filters envelope analysis. For the identification of AS and MR the frequency content of S1 and S2 is considered. The results from the bank of filters show that for aortic stenosis the ratio of energy in the frequency band centred on 10 Hz to the energy in the frequency band centred on 50 Hz is greater for S2 than for S1. No such relationship has been found for mitral regurgitation.