Comparison of EEG pattern classification methods for brain-computer interfaces

N. S. Dias, M. Kamrunnahar, P. M. Mendes, Steven Schiff, J. H. Correia

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

14 Citations (Scopus)

Abstract

The aim of this study is to compare 2 EEG pattern classification methods towards the development of BCI. The methods are: (1) discriminant stepwise, and (2) Principal Component Analysis (PCA) -Linear Discriminant Analysis (LDA) joint method. Both methods use Fisher's LDA approach, but differ in the data dimensionality reduction procedure. Data were recorded from 3 male subjects 20-30 years old. Three runs per subject took place. The classification methods were tested in 240 trials per subject after merging all runs for the same subject. The mental tasks performed were feet, tongue, left hand and right hand movement imagery. In order to avoid previous assumptions on preferable channel locations and frequency ranges, 105 (21 electrodes×5 frequency ranges) electroencephalogram (EEG) features were extracted from the data. The best performance for each classification method was taken into account. The discriminant stepwise method showed better performance than the PCA based method. The classification error by the stepwise method varied between 31.73% and 38.5% for all subjects whereas the error range using the PCA based method was 39.42% to 54%.

Original languageEnglish (US)
Title of host publication29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07
Pages2540-2543
Number of pages4
DOIs
StatePublished - Dec 1 2007
Event29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07 - Lyon, France
Duration: Aug 23 2007Aug 26 2007

Other

Other29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07
CountryFrance
CityLyon
Period8/23/078/26/07

Fingerprint

Brain-Computer Interfaces
Brain computer interface
Electroencephalography
Principal component analysis
Pattern recognition
Discriminant analysis
Merging
Data reduction
Principal Component Analysis
Discriminant Analysis
Hand
Imagery (Psychotherapy)
Tongue
Foot
Joints

All Science Journal Classification (ASJC) codes

  • Signal Processing
  • Biomedical Engineering
  • Computer Vision and Pattern Recognition
  • Health Informatics

Cite this

Dias, N. S., Kamrunnahar, M., Mendes, P. M., Schiff, S., & Correia, J. H. (2007). Comparison of EEG pattern classification methods for brain-computer interfaces. In 29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07 (pp. 2540-2543). [4352846] https://doi.org/10.1109/IEMBS.2007.4352846
Dias, N. S. ; Kamrunnahar, M. ; Mendes, P. M. ; Schiff, Steven ; Correia, J. H. / Comparison of EEG pattern classification methods for brain-computer interfaces. 29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07. 2007. pp. 2540-2543
@inproceedings{5a32616d1e37486f9635e94ea1cbb0c1,
title = "Comparison of EEG pattern classification methods for brain-computer interfaces",
abstract = "The aim of this study is to compare 2 EEG pattern classification methods towards the development of BCI. The methods are: (1) discriminant stepwise, and (2) Principal Component Analysis (PCA) -Linear Discriminant Analysis (LDA) joint method. Both methods use Fisher's LDA approach, but differ in the data dimensionality reduction procedure. Data were recorded from 3 male subjects 20-30 years old. Three runs per subject took place. The classification methods were tested in 240 trials per subject after merging all runs for the same subject. The mental tasks performed were feet, tongue, left hand and right hand movement imagery. In order to avoid previous assumptions on preferable channel locations and frequency ranges, 105 (21 electrodes×5 frequency ranges) electroencephalogram (EEG) features were extracted from the data. The best performance for each classification method was taken into account. The discriminant stepwise method showed better performance than the PCA based method. The classification error by the stepwise method varied between 31.73{\%} and 38.5{\%} for all subjects whereas the error range using the PCA based method was 39.42{\%} to 54{\%}.",
author = "Dias, {N. S.} and M. Kamrunnahar and Mendes, {P. M.} and Steven Schiff and Correia, {J. H.}",
year = "2007",
month = "12",
day = "1",
doi = "10.1109/IEMBS.2007.4352846",
language = "English (US)",
isbn = "1424407885",
pages = "2540--2543",
booktitle = "29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07",

}

Dias, NS, Kamrunnahar, M, Mendes, PM, Schiff, S & Correia, JH 2007, Comparison of EEG pattern classification methods for brain-computer interfaces. in 29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07., 4352846, pp. 2540-2543, 29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07, Lyon, France, 8/23/07. https://doi.org/10.1109/IEMBS.2007.4352846

Comparison of EEG pattern classification methods for brain-computer interfaces. / Dias, N. S.; Kamrunnahar, M.; Mendes, P. M.; Schiff, Steven; Correia, J. H.

29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07. 2007. p. 2540-2543 4352846.

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

TY - GEN

T1 - Comparison of EEG pattern classification methods for brain-computer interfaces

AU - Dias, N. S.

AU - Kamrunnahar, M.

AU - Mendes, P. M.

AU - Schiff, Steven

AU - Correia, J. H.

PY - 2007/12/1

Y1 - 2007/12/1

N2 - The aim of this study is to compare 2 EEG pattern classification methods towards the development of BCI. The methods are: (1) discriminant stepwise, and (2) Principal Component Analysis (PCA) -Linear Discriminant Analysis (LDA) joint method. Both methods use Fisher's LDA approach, but differ in the data dimensionality reduction procedure. Data were recorded from 3 male subjects 20-30 years old. Three runs per subject took place. The classification methods were tested in 240 trials per subject after merging all runs for the same subject. The mental tasks performed were feet, tongue, left hand and right hand movement imagery. In order to avoid previous assumptions on preferable channel locations and frequency ranges, 105 (21 electrodes×5 frequency ranges) electroencephalogram (EEG) features were extracted from the data. The best performance for each classification method was taken into account. The discriminant stepwise method showed better performance than the PCA based method. The classification error by the stepwise method varied between 31.73% and 38.5% for all subjects whereas the error range using the PCA based method was 39.42% to 54%.

AB - The aim of this study is to compare 2 EEG pattern classification methods towards the development of BCI. The methods are: (1) discriminant stepwise, and (2) Principal Component Analysis (PCA) -Linear Discriminant Analysis (LDA) joint method. Both methods use Fisher's LDA approach, but differ in the data dimensionality reduction procedure. Data were recorded from 3 male subjects 20-30 years old. Three runs per subject took place. The classification methods were tested in 240 trials per subject after merging all runs for the same subject. The mental tasks performed were feet, tongue, left hand and right hand movement imagery. In order to avoid previous assumptions on preferable channel locations and frequency ranges, 105 (21 electrodes×5 frequency ranges) electroencephalogram (EEG) features were extracted from the data. The best performance for each classification method was taken into account. The discriminant stepwise method showed better performance than the PCA based method. The classification error by the stepwise method varied between 31.73% and 38.5% for all subjects whereas the error range using the PCA based method was 39.42% to 54%.

UR - http://www.scopus.com/inward/record.url?scp=57649220811&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=57649220811&partnerID=8YFLogxK

U2 - 10.1109/IEMBS.2007.4352846

DO - 10.1109/IEMBS.2007.4352846

M3 - Conference contribution

C2 - 18002512

AN - SCOPUS:57649220811

SN - 1424407885

SN - 9781424407880

SP - 2540

EP - 2543

BT - 29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07

ER -

Dias NS, Kamrunnahar M, Mendes PM, Schiff S, Correia JH. Comparison of EEG pattern classification methods for brain-computer interfaces. In 29th Annual International Conference of IEEE-EMBS, Engineering in Medicine and Biology Society, EMBC'07. 2007. p. 2540-2543. 4352846 https://doi.org/10.1109/IEMBS.2007.4352846