A dynamic channel selection strategy for dense-Array ERP classification

Srinivas Kota, Lalit Gupta, Dennis L. Molfese, Ravi Vaidyanathan

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

The goal of this paper is to introduce a new strategy to accurately classify event-related potentials (ERPs), recorded using dense electrode arrays, into predefined brain activity categories. The challenge is to exploit the enhanced spatial information offered by dense arrays while overcoming the significant increase in the dimensionality problem introduced by the large increase in the number of channels. These conflicting objectives are achieved by introducing a spatiotemporal-array model to observe the dense-array ERP amplitude variations across channels and time, simultaneously. To account for latency variations and EEG noise in the array elements, each spatiotemporal element in the array is initially modeled as a Gaussian random variable. A two-step process that uses the Kolmogrov-Smirnov test and the Lilliefors test is formulated to select the array elements that have different Gaussian densities across all ERP categories. Selecting spatiotemporal elements that fit the assumed model and also statistically differ across the ERP categories not only ensures high classification accuracies but also decreases the dimensionality significantly. The selection is dynamic in the sense that selecting spatiotemporal-array elements corresponds to selecting ERP samples of different channels at different time instants. Each selected array element is classified using a univariate Gaussian classifier, and the resulting decisions are fused into a decision fusion vector that is classified using a discrete Bayes classifier. By converting an inherently multivariate classification problem into a simpler problem involving only univariate classifications, the dimensionality problem that plagues the design of practical multivariate ERP classifiers is circumvented. Consequently, classifiers can be designed to classify the ERPs that are unique to an individual without having to collect a prohibitively large ERP dataset from him/her. The application of the resulting dynamic-channel- selection-based classification strategy is demonstrated by designing and testing classifiers for eight subjects using ERPs from a Stroop color test and it is shown that the strategy yields high classification accuracies. Finally, it is noted that because of the generalized formulation of the strategy, it can be applied to various other problems involving the classification of multivariate signals acquired from multiple identical or multiple heterogeneous sensors.

Original languageEnglish (US)
Article number4663622
Pages (from-to)1040-1051
Number of pages12
JournalIEEE Transactions on Biomedical Engineering
Volume56
Issue number4
DOIs
StatePublished - Apr 1 2009

Fingerprint

Classifiers
Electroencephalography
Random variables
Brain
Fusion reactions
Color
Electrodes
Sensors
Testing

Keywords

  • Decision fusion
  • Dense electrode arrays
  • Dimensionality reduction
  • Dynamic channel selection
  • Event-related potentials (ERPs)
  • Spatiotemporal modeling

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

A dynamic channel selection strategy for dense-Array ERP classification. / Kota, Srinivas; Gupta, Lalit; Molfese, Dennis L.; Vaidyanathan, Ravi.

In: IEEE Transactions on Biomedical Engineering, Vol. 56, No. 4, 4663622, 01.04.2009, p. 1040-1051.

Research output: Contribution to journalArticle

Kota, Srinivas ; Gupta, Lalit ; Molfese, Dennis L. ; Vaidyanathan, Ravi. / A dynamic channel selection strategy for dense-Array ERP classification. In: IEEE Transactions on Biomedical Engineering. 2009 ; Vol. 56, No. 4. pp. 1040-1051.
@article{9424d3d86a104347839fb7464b6e7f17,
title = "A dynamic channel selection strategy for dense-Array ERP classification",
abstract = "The goal of this paper is to introduce a new strategy to accurately classify event-related potentials (ERPs), recorded using dense electrode arrays, into predefined brain activity categories. The challenge is to exploit the enhanced spatial information offered by dense arrays while overcoming the significant increase in the dimensionality problem introduced by the large increase in the number of channels. These conflicting objectives are achieved by introducing a spatiotemporal-array model to observe the dense-array ERP amplitude variations across channels and time, simultaneously. To account for latency variations and EEG noise in the array elements, each spatiotemporal element in the array is initially modeled as a Gaussian random variable. A two-step process that uses the Kolmogrov-Smirnov test and the Lilliefors test is formulated to select the array elements that have different Gaussian densities across all ERP categories. Selecting spatiotemporal elements that fit the assumed model and also statistically differ across the ERP categories not only ensures high classification accuracies but also decreases the dimensionality significantly. The selection is dynamic in the sense that selecting spatiotemporal-array elements corresponds to selecting ERP samples of different channels at different time instants. Each selected array element is classified using a univariate Gaussian classifier, and the resulting decisions are fused into a decision fusion vector that is classified using a discrete Bayes classifier. By converting an inherently multivariate classification problem into a simpler problem involving only univariate classifications, the dimensionality problem that plagues the design of practical multivariate ERP classifiers is circumvented. Consequently, classifiers can be designed to classify the ERPs that are unique to an individual without having to collect a prohibitively large ERP dataset from him/her. The application of the resulting dynamic-channel- selection-based classification strategy is demonstrated by designing and testing classifiers for eight subjects using ERPs from a Stroop color test and it is shown that the strategy yields high classification accuracies. Finally, it is noted that because of the generalized formulation of the strategy, it can be applied to various other problems involving the classification of multivariate signals acquired from multiple identical or multiple heterogeneous sensors.",
keywords = "Decision fusion, Dense electrode arrays, Dimensionality reduction, Dynamic channel selection, Event-related potentials (ERPs), Spatiotemporal modeling",
author = "Srinivas Kota and Lalit Gupta and Molfese, {Dennis L.} and Ravi Vaidyanathan",
year = "2009",
month = "4",
day = "1",
doi = "10.1109/TBME.2008.2006985",
language = "English (US)",
volume = "56",
pages = "1040--1051",
journal = "IEEE Transactions on Biomedical Engineering",
issn = "0018-9294",
publisher = "IEEE Computer Society",
number = "4",

}

TY - JOUR

T1 - A dynamic channel selection strategy for dense-Array ERP classification

AU - Kota, Srinivas

AU - Gupta, Lalit

AU - Molfese, Dennis L.

AU - Vaidyanathan, Ravi

PY - 2009/4/1

Y1 - 2009/4/1

N2 - The goal of this paper is to introduce a new strategy to accurately classify event-related potentials (ERPs), recorded using dense electrode arrays, into predefined brain activity categories. The challenge is to exploit the enhanced spatial information offered by dense arrays while overcoming the significant increase in the dimensionality problem introduced by the large increase in the number of channels. These conflicting objectives are achieved by introducing a spatiotemporal-array model to observe the dense-array ERP amplitude variations across channels and time, simultaneously. To account for latency variations and EEG noise in the array elements, each spatiotemporal element in the array is initially modeled as a Gaussian random variable. A two-step process that uses the Kolmogrov-Smirnov test and the Lilliefors test is formulated to select the array elements that have different Gaussian densities across all ERP categories. Selecting spatiotemporal elements that fit the assumed model and also statistically differ across the ERP categories not only ensures high classification accuracies but also decreases the dimensionality significantly. The selection is dynamic in the sense that selecting spatiotemporal-array elements corresponds to selecting ERP samples of different channels at different time instants. Each selected array element is classified using a univariate Gaussian classifier, and the resulting decisions are fused into a decision fusion vector that is classified using a discrete Bayes classifier. By converting an inherently multivariate classification problem into a simpler problem involving only univariate classifications, the dimensionality problem that plagues the design of practical multivariate ERP classifiers is circumvented. Consequently, classifiers can be designed to classify the ERPs that are unique to an individual without having to collect a prohibitively large ERP dataset from him/her. The application of the resulting dynamic-channel- selection-based classification strategy is demonstrated by designing and testing classifiers for eight subjects using ERPs from a Stroop color test and it is shown that the strategy yields high classification accuracies. Finally, it is noted that because of the generalized formulation of the strategy, it can be applied to various other problems involving the classification of multivariate signals acquired from multiple identical or multiple heterogeneous sensors.

AB - The goal of this paper is to introduce a new strategy to accurately classify event-related potentials (ERPs), recorded using dense electrode arrays, into predefined brain activity categories. The challenge is to exploit the enhanced spatial information offered by dense arrays while overcoming the significant increase in the dimensionality problem introduced by the large increase in the number of channels. These conflicting objectives are achieved by introducing a spatiotemporal-array model to observe the dense-array ERP amplitude variations across channels and time, simultaneously. To account for latency variations and EEG noise in the array elements, each spatiotemporal element in the array is initially modeled as a Gaussian random variable. A two-step process that uses the Kolmogrov-Smirnov test and the Lilliefors test is formulated to select the array elements that have different Gaussian densities across all ERP categories. Selecting spatiotemporal elements that fit the assumed model and also statistically differ across the ERP categories not only ensures high classification accuracies but also decreases the dimensionality significantly. The selection is dynamic in the sense that selecting spatiotemporal-array elements corresponds to selecting ERP samples of different channels at different time instants. Each selected array element is classified using a univariate Gaussian classifier, and the resulting decisions are fused into a decision fusion vector that is classified using a discrete Bayes classifier. By converting an inherently multivariate classification problem into a simpler problem involving only univariate classifications, the dimensionality problem that plagues the design of practical multivariate ERP classifiers is circumvented. Consequently, classifiers can be designed to classify the ERPs that are unique to an individual without having to collect a prohibitively large ERP dataset from him/her. The application of the resulting dynamic-channel- selection-based classification strategy is demonstrated by designing and testing classifiers for eight subjects using ERPs from a Stroop color test and it is shown that the strategy yields high classification accuracies. Finally, it is noted that because of the generalized formulation of the strategy, it can be applied to various other problems involving the classification of multivariate signals acquired from multiple identical or multiple heterogeneous sensors.

KW - Decision fusion

KW - Dense electrode arrays

KW - Dimensionality reduction

KW - Dynamic channel selection

KW - Event-related potentials (ERPs)

KW - Spatiotemporal modeling

UR - http://www.scopus.com/inward/record.url?scp=67149143328&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=67149143328&partnerID=8YFLogxK

U2 - 10.1109/TBME.2008.2006985

DO - 10.1109/TBME.2008.2006985

M3 - Article

C2 - 19272892

AN - SCOPUS:67149143328

VL - 56

SP - 1040

EP - 1051

JO - IEEE Transactions on Biomedical Engineering

JF - IEEE Transactions on Biomedical Engineering

SN - 0018-9294

IS - 4

M1 - 4663622

ER -