Title: Classification of the heart sounds via artificial neural network

Authors: Gur Emre Guraksin, Ucman Ergun, Omer Deperlioglu

Addresses: Department of Computer Engineering, Afyon Kocatepe University, Campus of Ahmet Necdet Sezer, Faculty of Engineering, Afyonkarahisar 03200, Turkey. ' Department of Biomedical Engineering, Afyon Kocatepe University, Campus of Ahmet Necdet Sezer, Faculty of Engineering, Afyonkarahisar 03200, Turkey. ' Department of Biomedical Engineering, Afyon Kocatepe University, Campus of Ahmet Necdet Sezer, Faculty of Engineering, Afyonkarahisar 03200, Turkey

Abstract: Auscultation with stethoscope is a preferential method that the doctors use in order to differentiate normal cardiac systems from the abnormal ones that come out. On the other hand, the method of auscultation with stethoscope requires medical expert experience and careful listening. Because of the problems that can be faced, listening process with stethoscope, that is auscultation, falls behind in the search of the heart abnormalities. In this study, the frequency analysis of the heart sounds taken by the electronic stethoscope is implemented via a pocket computer. The parameters gained from the frequency analysis are again classified with the neural network on the pocket computer. Thus, hearing the heart sounds, medical doctors at the same time will be able to follow the information gained through the frequency and artificial neural network analysis on the pocket computer so that they will be able to make more accurate diagnoses.

Keywords: heart sounds; discrete Fourier transform; DFT; artificial neural networks; ANNs; cardiac systems; heart abnormalities; electronic stethoscopes; frequency analysis; pocket computers; medical diagnosis; healthcare technology; electronic healthcare; e-healthcare; mobile devices.

DOI: 10.1504/IJRIS.2010.036874

International Journal of Reasoning-based Intelligent Systems, 2010 Vol.2 No.3/4, pp.272 - 278

Published online: 12 Nov 2010 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article