Abstract

This article describes a sophisticated method for detection of Heart valve defects at early stage i.e Grade 1 murmur without the expertise of a doctor. The diagnosis based on heard heart sounds through, either a conventional acoustic or an electronic stethoscope is itself a very specialized skill that will take years to acquire. Sometimes the doctors cannot detect these defects till the murmurs reach grade 3 and above which generally is too late for prognosis. Here, we have taken the recorded heart sounds from 350 subjects and performed the Fast Fourier Transform(FFT) on it, but it didn’t give satisfying result. We have also recoded heart sounds using phonograph for 20 subjects in noise free environment. In this technique Frequency component with the maximum magnitude (in Hertz) was observed to be of varying values across some heart sounds (for e.g., heart sound from subject 41=13.0588Hz and heart sound from subject 58=324.5293Hz). Hence normal heart sounds could not be categorized in a generic way. To overcome this problem, we have used Shannon energy method on same data file, which will classify the condition of heart by finding S1(lub) and S2(dub) frequency component, if they lie between 30-100Hz, the heart is normal and if it is above 100Hz then the heart function is abnormal..

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call