Optimal channel selection based on statistical analysis in high dimensional NIRS data

Min Ho Lee, Siamac Fazli, Seong Whan Lee

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Near-infrared spectroscopy (NIRS) is an optical imaging method that has recently been investigated for non-invasive Brain Computer Interfaces (BCI). The performance of NIRS-based BCI can deteriorate when the number of channels becomes larger. Here we present three types of channel selection methods based on ranked channels, pre-defined channel configurations and statistical analysis for high dimensional NIRS data. The optimal combination of channels is selected by the highest classification accuracy rate based on Linear Discriminant Analysis (LDA). Experimental results show that the three considered types of channel selection methods achieve higher classification performance by removing the noisy and non-informative channels. Also the proposed statistical channel selection method can reduce the computation time significantly without any loss of classification accuracy.

Original languageEnglish
Title of host publication2013 International Winter Workshop on Brain-Computer Interface, BCI 2013
Pages95-97
Number of pages3
DOIs
Publication statusPublished - May 17 2013
Externally publishedYes
Event2013 International Winter Workshop on Brain-Computer Interface, BCI 2013 - Gangwon Province, Korea, Republic of
Duration: Feb 18 2013Feb 20 2013

Publication series

Name2013 International Winter Workshop on Brain-Computer Interface, BCI 2013

Other

Other2013 International Winter Workshop on Brain-Computer Interface, BCI 2013
CountryKorea, Republic of
CityGangwon Province
Period2/18/132/20/13

Fingerprint

Near infrared spectroscopy
Statistical methods
Brain computer interface
Discriminant analysis
Imaging techniques

Keywords

  • NIRS-based BCI
  • Optimal channel selection
  • Statistical channel selection

ASJC Scopus subject areas

  • Human-Computer Interaction

Cite this

Lee, M. H., Fazli, S., & Lee, S. W. (2013). Optimal channel selection based on statistical analysis in high dimensional NIRS data. In 2013 International Winter Workshop on Brain-Computer Interface, BCI 2013 (pp. 95-97). [6506643] (2013 International Winter Workshop on Brain-Computer Interface, BCI 2013). https://doi.org/10.1109/IWW-BCI.2013.6506643

Optimal channel selection based on statistical analysis in high dimensional NIRS data. / Lee, Min Ho; Fazli, Siamac; Lee, Seong Whan.

2013 International Winter Workshop on Brain-Computer Interface, BCI 2013. 2013. p. 95-97 6506643 (2013 International Winter Workshop on Brain-Computer Interface, BCI 2013).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Lee, MH, Fazli, S & Lee, SW 2013, Optimal channel selection based on statistical analysis in high dimensional NIRS data. in 2013 International Winter Workshop on Brain-Computer Interface, BCI 2013., 6506643, 2013 International Winter Workshop on Brain-Computer Interface, BCI 2013, pp. 95-97, 2013 International Winter Workshop on Brain-Computer Interface, BCI 2013, Gangwon Province, Korea, Republic of, 2/18/13. https://doi.org/10.1109/IWW-BCI.2013.6506643
Lee MH, Fazli S, Lee SW. Optimal channel selection based on statistical analysis in high dimensional NIRS data. In 2013 International Winter Workshop on Brain-Computer Interface, BCI 2013. 2013. p. 95-97. 6506643. (2013 International Winter Workshop on Brain-Computer Interface, BCI 2013). https://doi.org/10.1109/IWW-BCI.2013.6506643
Lee, Min Ho ; Fazli, Siamac ; Lee, Seong Whan. / Optimal channel selection based on statistical analysis in high dimensional NIRS data. 2013 International Winter Workshop on Brain-Computer Interface, BCI 2013. 2013. pp. 95-97 (2013 International Winter Workshop on Brain-Computer Interface, BCI 2013).
@inproceedings{bcd36fd18eb343eea584c345c3243b15,
title = "Optimal channel selection based on statistical analysis in high dimensional NIRS data",
abstract = "Near-infrared spectroscopy (NIRS) is an optical imaging method that has recently been investigated for non-invasive Brain Computer Interfaces (BCI). The performance of NIRS-based BCI can deteriorate when the number of channels becomes larger. Here we present three types of channel selection methods based on ranked channels, pre-defined channel configurations and statistical analysis for high dimensional NIRS data. The optimal combination of channels is selected by the highest classification accuracy rate based on Linear Discriminant Analysis (LDA). Experimental results show that the three considered types of channel selection methods achieve higher classification performance by removing the noisy and non-informative channels. Also the proposed statistical channel selection method can reduce the computation time significantly without any loss of classification accuracy.",
keywords = "NIRS-based BCI, Optimal channel selection, Statistical channel selection",
author = "Lee, {Min Ho} and Siamac Fazli and Lee, {Seong Whan}",
year = "2013",
month = "5",
day = "17",
doi = "10.1109/IWW-BCI.2013.6506643",
language = "English",
isbn = "9781467359733",
series = "2013 International Winter Workshop on Brain-Computer Interface, BCI 2013",
pages = "95--97",
booktitle = "2013 International Winter Workshop on Brain-Computer Interface, BCI 2013",

}

TY - GEN

T1 - Optimal channel selection based on statistical analysis in high dimensional NIRS data

AU - Lee, Min Ho

AU - Fazli, Siamac

AU - Lee, Seong Whan

PY - 2013/5/17

Y1 - 2013/5/17

N2 - Near-infrared spectroscopy (NIRS) is an optical imaging method that has recently been investigated for non-invasive Brain Computer Interfaces (BCI). The performance of NIRS-based BCI can deteriorate when the number of channels becomes larger. Here we present three types of channel selection methods based on ranked channels, pre-defined channel configurations and statistical analysis for high dimensional NIRS data. The optimal combination of channels is selected by the highest classification accuracy rate based on Linear Discriminant Analysis (LDA). Experimental results show that the three considered types of channel selection methods achieve higher classification performance by removing the noisy and non-informative channels. Also the proposed statistical channel selection method can reduce the computation time significantly without any loss of classification accuracy.

AB - Near-infrared spectroscopy (NIRS) is an optical imaging method that has recently been investigated for non-invasive Brain Computer Interfaces (BCI). The performance of NIRS-based BCI can deteriorate when the number of channels becomes larger. Here we present three types of channel selection methods based on ranked channels, pre-defined channel configurations and statistical analysis for high dimensional NIRS data. The optimal combination of channels is selected by the highest classification accuracy rate based on Linear Discriminant Analysis (LDA). Experimental results show that the three considered types of channel selection methods achieve higher classification performance by removing the noisy and non-informative channels. Also the proposed statistical channel selection method can reduce the computation time significantly without any loss of classification accuracy.

KW - NIRS-based BCI

KW - Optimal channel selection

KW - Statistical channel selection

UR - http://www.scopus.com/inward/record.url?scp=84877713605&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84877713605&partnerID=8YFLogxK

U2 - 10.1109/IWW-BCI.2013.6506643

DO - 10.1109/IWW-BCI.2013.6506643

M3 - Conference contribution

SN - 9781467359733

T3 - 2013 International Winter Workshop on Brain-Computer Interface, BCI 2013

SP - 95

EP - 97

BT - 2013 International Winter Workshop on Brain-Computer Interface, BCI 2013

ER -