Classification of heart sounds by using wavelet transform

Author

Say, Özgür ; Dokur, Zümray ; Ölmez, Tamer

Author_Institution

Dept. of Electron. & Commun. Eng., Istanbul Tech. Univ., Turkey

Volume

1

fYear

2002

fDate

2002

Firstpage

128

Abstract

A novel method is presented to determine the features of heart sounds. Feature vectors are formed by using the wavelet detail and approximation coefficients at the second and the sixth decomposition levels, respectively. Decision making is performed in four stages: Segmentation of the first and second heart sounds, normalization process, feature extraction, and classification by the artificial neural network. In this study, nine different heart sounds are classified.

Keywords

acoustic signal processing; cardiology; digital simulation; feature extraction; medical signal processing; spectral analysis; vectors; wavelet transforms; aortic regurgitation; aortic stenosis; approximation coefficients; artificial neural network; decision making; grow and learn neural network; heart sounds classification; medical diagnostic technique; mitral regurgitation; mitral stenosis; normal heart sound; patent ductus arteriosis; pulmonary stenosis; second decomposition level; sixth decomposition level; summation gallop; ventricular septal defect; wavelet detail; Acoustical engineering; Artificial neural networks; Data mining; Decision making; Discrete wavelet transforms; Feature extraction; Heart; Signal processing; Wavelet coefficients; Wavelet transforms;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology, 2002. 24th Annual Conference and the Annual Fall Meeting of the Biomedical Engineering Society EMBS/BMES Conference, 2002. Proceedings of the Second Joint

ISSN

1094-687X

Print_ISBN

0-7803-7612-9

Type

conf

DOI

10.1109/IEMBS.2002.1134419

Filename

1134419