Model comparison for automatic characterization and classification of average ERPs using visual oddball paradigm

A. C. Merzagora, M. Butti, R. Polikar, M. Izzetoglu, S. Bunce, S. Cerutti, A. M. Bianchi, B. Onaral

Research output: Contribution to journalArticle

11 Citations (Scopus)

Abstract

Objective: To determine whether automated classifiers can be used for correctly identifying target categorization responses from averaged event-related potentials (ERPs) along with identifying appropriate features and classification models for computer-assisted investigation of attentional processes. Methods: ERPs were recorded during a target categorization task. Automated classification of average target ERPs versus average non-target ERPs was performed by extracting different combinations of features from the P300 and N200 components, which were used to train six classifiers: Euclidean classifier (EC), Mahalanobis discriminant (MD), quadratic classifier (QC), Fisher linear discriminant (FLD), multi-layer perceptron neural network (MLP) and support vector machine (SVM). Results: The best classification performance (accuracy: 91-92%; sensitivity: 85-86%; specificity: 95-99%) was provided by QC, MLP, SVM on feature vectors extracted from P300 recorded at multiple sites. In general, non-linear and non-parametric classifiers (QC, MLP, SVM) performed better than linear classifiers (EC, MD, FLD). The N200 did not explain variance beyond that of P300 recorded at multiple sites. Conclusions: The results suggest that automatic characterization and classification of average target and non-target ERPs is feasible. Features of P300 recorded at multiple sites used to train non-linear classifiers are recommended for optimal classification performance. Significance: Automatic characterization of target ERPs can provide an objective approach for detecting and diagnosing abnormalities and evaluating interventions for clinical populations, paving the way for future real-time monitoring of attentional processes.

Original languageEnglish (US)
Pages (from-to)264-274
Number of pages11
JournalClinical Neurophysiology
Volume120
Issue number2
DOIs
StatePublished - Feb 1 2009

Fingerprint

Evoked Potentials
P300 Event-Related Potentials
Neural Networks (Computer)
Computer Simulation
Population
Support Vector Machine

All Science Journal Classification (ASJC) codes

  • Sensory Systems
  • Neurology
  • Clinical Neurology
  • Physiology (medical)

Cite this

Merzagora, A. C. ; Butti, M. ; Polikar, R. ; Izzetoglu, M. ; Bunce, S. ; Cerutti, S. ; Bianchi, A. M. ; Onaral, B. / Model comparison for automatic characterization and classification of average ERPs using visual oddball paradigm. In: Clinical Neurophysiology. 2009 ; Vol. 120, No. 2. pp. 264-274.
@article{73424bad953f4c8183d23ce265440ea1,
title = "Model comparison for automatic characterization and classification of average ERPs using visual oddball paradigm",
abstract = "Objective: To determine whether automated classifiers can be used for correctly identifying target categorization responses from averaged event-related potentials (ERPs) along with identifying appropriate features and classification models for computer-assisted investigation of attentional processes. Methods: ERPs were recorded during a target categorization task. Automated classification of average target ERPs versus average non-target ERPs was performed by extracting different combinations of features from the P300 and N200 components, which were used to train six classifiers: Euclidean classifier (EC), Mahalanobis discriminant (MD), quadratic classifier (QC), Fisher linear discriminant (FLD), multi-layer perceptron neural network (MLP) and support vector machine (SVM). Results: The best classification performance (accuracy: 91-92{\%}; sensitivity: 85-86{\%}; specificity: 95-99{\%}) was provided by QC, MLP, SVM on feature vectors extracted from P300 recorded at multiple sites. In general, non-linear and non-parametric classifiers (QC, MLP, SVM) performed better than linear classifiers (EC, MD, FLD). The N200 did not explain variance beyond that of P300 recorded at multiple sites. Conclusions: The results suggest that automatic characterization and classification of average target and non-target ERPs is feasible. Features of P300 recorded at multiple sites used to train non-linear classifiers are recommended for optimal classification performance. Significance: Automatic characterization of target ERPs can provide an objective approach for detecting and diagnosing abnormalities and evaluating interventions for clinical populations, paving the way for future real-time monitoring of attentional processes.",
author = "Merzagora, {A. C.} and M. Butti and R. Polikar and M. Izzetoglu and S. Bunce and S. Cerutti and Bianchi, {A. M.} and B. Onaral",
year = "2009",
month = "2",
day = "1",
doi = "10.1016/j.clinph.2008.10.016",
language = "English (US)",
volume = "120",
pages = "264--274",
journal = "Clinical Neurophysiology",
issn = "1388-2457",
publisher = "Elsevier Ireland Ltd",
number = "2",

}

Model comparison for automatic characterization and classification of average ERPs using visual oddball paradigm. / Merzagora, A. C.; Butti, M.; Polikar, R.; Izzetoglu, M.; Bunce, S.; Cerutti, S.; Bianchi, A. M.; Onaral, B.

In: Clinical Neurophysiology, Vol. 120, No. 2, 01.02.2009, p. 264-274.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Model comparison for automatic characterization and classification of average ERPs using visual oddball paradigm

AU - Merzagora, A. C.

AU - Butti, M.

AU - Polikar, R.

AU - Izzetoglu, M.

AU - Bunce, S.

AU - Cerutti, S.

AU - Bianchi, A. M.

AU - Onaral, B.

PY - 2009/2/1

Y1 - 2009/2/1

N2 - Objective: To determine whether automated classifiers can be used for correctly identifying target categorization responses from averaged event-related potentials (ERPs) along with identifying appropriate features and classification models for computer-assisted investigation of attentional processes. Methods: ERPs were recorded during a target categorization task. Automated classification of average target ERPs versus average non-target ERPs was performed by extracting different combinations of features from the P300 and N200 components, which were used to train six classifiers: Euclidean classifier (EC), Mahalanobis discriminant (MD), quadratic classifier (QC), Fisher linear discriminant (FLD), multi-layer perceptron neural network (MLP) and support vector machine (SVM). Results: The best classification performance (accuracy: 91-92%; sensitivity: 85-86%; specificity: 95-99%) was provided by QC, MLP, SVM on feature vectors extracted from P300 recorded at multiple sites. In general, non-linear and non-parametric classifiers (QC, MLP, SVM) performed better than linear classifiers (EC, MD, FLD). The N200 did not explain variance beyond that of P300 recorded at multiple sites. Conclusions: The results suggest that automatic characterization and classification of average target and non-target ERPs is feasible. Features of P300 recorded at multiple sites used to train non-linear classifiers are recommended for optimal classification performance. Significance: Automatic characterization of target ERPs can provide an objective approach for detecting and diagnosing abnormalities and evaluating interventions for clinical populations, paving the way for future real-time monitoring of attentional processes.

AB - Objective: To determine whether automated classifiers can be used for correctly identifying target categorization responses from averaged event-related potentials (ERPs) along with identifying appropriate features and classification models for computer-assisted investigation of attentional processes. Methods: ERPs were recorded during a target categorization task. Automated classification of average target ERPs versus average non-target ERPs was performed by extracting different combinations of features from the P300 and N200 components, which were used to train six classifiers: Euclidean classifier (EC), Mahalanobis discriminant (MD), quadratic classifier (QC), Fisher linear discriminant (FLD), multi-layer perceptron neural network (MLP) and support vector machine (SVM). Results: The best classification performance (accuracy: 91-92%; sensitivity: 85-86%; specificity: 95-99%) was provided by QC, MLP, SVM on feature vectors extracted from P300 recorded at multiple sites. In general, non-linear and non-parametric classifiers (QC, MLP, SVM) performed better than linear classifiers (EC, MD, FLD). The N200 did not explain variance beyond that of P300 recorded at multiple sites. Conclusions: The results suggest that automatic characterization and classification of average target and non-target ERPs is feasible. Features of P300 recorded at multiple sites used to train non-linear classifiers are recommended for optimal classification performance. Significance: Automatic characterization of target ERPs can provide an objective approach for detecting and diagnosing abnormalities and evaluating interventions for clinical populations, paving the way for future real-time monitoring of attentional processes.

UR - http://www.scopus.com/inward/record.url?scp=59149106362&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=59149106362&partnerID=8YFLogxK

U2 - 10.1016/j.clinph.2008.10.016

DO - 10.1016/j.clinph.2008.10.016

M3 - Article

C2 - 19062338

AN - SCOPUS:59149106362

VL - 120

SP - 264

EP - 274

JO - Clinical Neurophysiology

JF - Clinical Neurophysiology

SN - 1388-2457

IS - 2

ER -