Respiratory sounds classification using cepstral analysis and Gaussian mixture models

Author

Bahoura, M. ; Pelletier, C.

Author_Institution

Departement de Mathematiques, Universite du Quebec a Rimouski, Que., Canada

Volume

1

fYear

2004

fDate

1-5 Sept. 2004

Firstpage

9

Lastpage

12

Abstract

The cepstral analysis is proposed with Gaussian mixture models (GMM) method to classify respiratory sounds in two categories: normal and wheezing. The sound signal is divided in overlapped segments, which are characterized by a reduced dimension feature vectors using Mel-frequency cepstral coefficients (MFCC) or subband based cepstral parameters (SBC). The proposed schema is compared with other classifiers: vector quantization (VQ) and multi-layer perceptron (MLP) neural networks. A post processing is proposed to improve the classification results.

Keywords

bioacoustics; cepstral analysis; medical signal processing; multilayer perceptrons; pneumodynamics; signal classification; vector quantisation; Gaussian mixture models; Mel-frequency cepstral coefficients; cepstral analysis; multilayer perceptron neural networks; normal sound; respiratory sounds classification; subband based cepstral parameters; vector quantization neural networks; wheezing; Cepstral analysis; Continuous wavelet transforms; Discrete wavelet transforms; Feature extraction; Filters; Fourier transforms; Mel frequency cepstral coefficient; Neural networks; Wavelet packets; Wavelet transforms; Gaussian mixture models; Respiratory sounds; cepstral analysis; wheezes;

fLanguage

English

Publisher

ieee

Conference_Titel

Engineering in Medicine and Biology Society, 2004. IEMBS '04. 26th Annual International Conference of the IEEE

Conference_Location

San Francisco, CA

Print_ISBN

0-7803-8439-3

Type

conf

DOI

10.1109/IEMBS.2004.1403077

Filename

1403077