Combining Wavelet Analysis and Bayesian Networks for the Classification of Auditory Brainstem Response

R Zhang, G McAllister, BW Scotney, SI McClean, G Houston

Research output: Contribution to journalArticle

27 Citations (Scopus)

Abstract

The auditory brainstem response (ABR) has become a routine clinical tool for hearing and neurological assessment. In order to pick out the ABR from the background EEG activity that obscures it, stimulus-synchronized averaging of many repeated trials is necessary, typically requiring up to 2000 repetitions. This number of repetitions can be very difficult, time consuming and uncomfortable for some subjects. In this study, a method combining wavelet analysis and Bayesian networks is introduced to reduce the required number of repetitions, which could offer a great advantage in the clinical situation. 314 ABRs with 64 repetitions and 155 ABRs with 128 repetitions recorded from eight subjects are used here. A wavelet transform is applied to each of the ABRs, and the important features of the ABRs are extracted by thresholding and matching the wavelet coefficients. The significant wavelet coefficients that represent the extracted features of the ABRs are then used as the variables to build the Bayesian network for classification of the ABRs. In order to estimate the performance of this approach, stratified ten-fold cross-validation is used.
LanguageEnglish
Pages458-467
JournalIEEE Transactions on Information Technology in BioMedicine
Volume10
Issue number3
DOIs
Publication statusPublished - 1 Jul 2006

Fingerprint

Wavelet Analysis
Brain Stem Auditory Evoked Potentials
Wavelet analysis
Bayesian networks
Audition
Electroencephalography
Wavelet transforms
Hearing

Keywords

  • Auditory brainstem response (ABR)
  • Bayesian networks
  • classification
  • stratified tenfold cross validation
  • wavelet analysis

Cite this

@article{621aae242f704004b71111102691080c,
title = "Combining Wavelet Analysis and Bayesian Networks for the Classification of Auditory Brainstem Response",
abstract = "The auditory brainstem response (ABR) has become a routine clinical tool for hearing and neurological assessment. In order to pick out the ABR from the background EEG activity that obscures it, stimulus-synchronized averaging of many repeated trials is necessary, typically requiring up to 2000 repetitions. This number of repetitions can be very difficult, time consuming and uncomfortable for some subjects. In this study, a method combining wavelet analysis and Bayesian networks is introduced to reduce the required number of repetitions, which could offer a great advantage in the clinical situation. 314 ABRs with 64 repetitions and 155 ABRs with 128 repetitions recorded from eight subjects are used here. A wavelet transform is applied to each of the ABRs, and the important features of the ABRs are extracted by thresholding and matching the wavelet coefficients. The significant wavelet coefficients that represent the extracted features of the ABRs are then used as the variables to build the Bayesian network for classification of the ABRs. In order to estimate the performance of this approach, stratified ten-fold cross-validation is used.",
keywords = "Auditory brainstem response (ABR), Bayesian networks, classification, stratified tenfold cross validation, wavelet analysis",
author = "R Zhang and G McAllister and BW Scotney and SI McClean and G Houston",
note = "Other Details ------------------------------------ This paper was selected (with re-review) for journal publication from papers presented at the 18th IEEE International Symposium on Computer-Based Medical Systems, June 2005. The paper develops an algorithm that enables hearing response to be assessed using less than one-tenth of the data (from auditory brainstem responses to stimuli) typically required by techniques currently used by audiologists. Thus the length of assessment sessions may be reduced, with corresponding reductions in patient discomfort and audiologists' time. Hearing classification accuracy has been evaluated in collaboration with (and using data provided by) an Audiological Scientist at the Royal Hospitals, Belfast (Glen Houston).",
year = "2006",
month = "7",
day = "1",
doi = "10.1109/TITB.2005.863865",
language = "English",
volume = "10",
pages = "458--467",
journal = "IEEE Transactions on Information Technology in BioMedicine",
issn = "1089-7771",
number = "3",

}

Combining Wavelet Analysis and Bayesian Networks for the Classification of Auditory Brainstem Response. / Zhang, R; McAllister, G; Scotney, BW; McClean, SI; Houston, G.

In: IEEE Transactions on Information Technology in BioMedicine, Vol. 10, No. 3, 01.07.2006, p. 458-467.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Combining Wavelet Analysis and Bayesian Networks for the Classification of Auditory Brainstem Response

AU - Zhang, R

AU - McAllister, G

AU - Scotney, BW

AU - McClean, SI

AU - Houston, G

N1 - Other Details ------------------------------------ This paper was selected (with re-review) for journal publication from papers presented at the 18th IEEE International Symposium on Computer-Based Medical Systems, June 2005. The paper develops an algorithm that enables hearing response to be assessed using less than one-tenth of the data (from auditory brainstem responses to stimuli) typically required by techniques currently used by audiologists. Thus the length of assessment sessions may be reduced, with corresponding reductions in patient discomfort and audiologists' time. Hearing classification accuracy has been evaluated in collaboration with (and using data provided by) an Audiological Scientist at the Royal Hospitals, Belfast (Glen Houston).

PY - 2006/7/1

Y1 - 2006/7/1

N2 - The auditory brainstem response (ABR) has become a routine clinical tool for hearing and neurological assessment. In order to pick out the ABR from the background EEG activity that obscures it, stimulus-synchronized averaging of many repeated trials is necessary, typically requiring up to 2000 repetitions. This number of repetitions can be very difficult, time consuming and uncomfortable for some subjects. In this study, a method combining wavelet analysis and Bayesian networks is introduced to reduce the required number of repetitions, which could offer a great advantage in the clinical situation. 314 ABRs with 64 repetitions and 155 ABRs with 128 repetitions recorded from eight subjects are used here. A wavelet transform is applied to each of the ABRs, and the important features of the ABRs are extracted by thresholding and matching the wavelet coefficients. The significant wavelet coefficients that represent the extracted features of the ABRs are then used as the variables to build the Bayesian network for classification of the ABRs. In order to estimate the performance of this approach, stratified ten-fold cross-validation is used.

AB - The auditory brainstem response (ABR) has become a routine clinical tool for hearing and neurological assessment. In order to pick out the ABR from the background EEG activity that obscures it, stimulus-synchronized averaging of many repeated trials is necessary, typically requiring up to 2000 repetitions. This number of repetitions can be very difficult, time consuming and uncomfortable for some subjects. In this study, a method combining wavelet analysis and Bayesian networks is introduced to reduce the required number of repetitions, which could offer a great advantage in the clinical situation. 314 ABRs with 64 repetitions and 155 ABRs with 128 repetitions recorded from eight subjects are used here. A wavelet transform is applied to each of the ABRs, and the important features of the ABRs are extracted by thresholding and matching the wavelet coefficients. The significant wavelet coefficients that represent the extracted features of the ABRs are then used as the variables to build the Bayesian network for classification of the ABRs. In order to estimate the performance of this approach, stratified ten-fold cross-validation is used.

KW - Auditory brainstem response (ABR)

KW - Bayesian networks

KW - classification

KW - stratified tenfold cross validation

KW - wavelet analysis

U2 - 10.1109/TITB.2005.863865

DO - 10.1109/TITB.2005.863865

M3 - Article

VL - 10

SP - 458

EP - 467

JO - IEEE Transactions on Information Technology in BioMedicine

T2 - IEEE Transactions on Information Technology in BioMedicine

JF - IEEE Transactions on Information Technology in BioMedicine

SN - 1089-7771

IS - 3

ER -