Detection of cochlear hearing loss applying wavelet packets and support vector machines

Author

Dietl, Hubert ; Weiss, Stephan

Author_Institution

Dept. of Electron. & Comput. Sci., Southampton Univ., UK

Volume

2

fYear

2004

fDate

7-10 Nov. 2004

Firstpage

1575

Abstract

The aim of this paper is to evaluate the application of the wavelet packet transform (WP) and support vector machines (SVM) to transient evoked otoacoustic emissions (TEOAE) in order to achieve a detection of frequency-specific hearing loss. We introduce a system to determine detection rates between groups of persons with normal hearing, high frequency hearing loss, and pantonal hearing loss. The validity and use of our approach is verified on a different patient group.

Keywords

acoustic signal detection; auditory evoked potentials; feature extraction; otoacoustic emissions; signal classification; support vector machines; wavelet transforms; SNR; SVM classifier; cochlear hearing loss detection; feature extraction; frequency-specific hearing loss; high frequency hearing loss; pantonal hearing loss; signal-to-noise ratio; support vector machines; transient evoked otoacoustic emissions; wavelet packet transform; Deafness; Discrete wavelet transforms; Feature extraction; Frequency; Support vector machine classification; Support vector machines; Testing; Training data; Wavelet packets; Wavelet transforms;

fLanguage

English

Publisher

ieee

Conference_Titel

Signals, Systems and Computers, 2004. Conference Record of the Thirty-Eighth Asilomar Conference on

Print_ISBN

0-7803-8622-1

Type

conf

DOI

10.1109/ACSSC.2004.1399421

Filename

1399421