Time-dependent common spatial patterns optimization for EEG signal classification

Tae Eui Kam, Seong Whan Lee

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

Abstract

Recognizing Event-Related Desynchronization or Synchronization (ERD/ERS) patterns generated by motor imagery tasks is an important process in Brain-computer interfaces (BCI). One of the most well-known algorithms to extract the discriminative patterns is Common Spatial Patterns (CSP). It finds an optimal spatial filter considering the spatial distribution of the ERD/ERS patterns. The CSP algorithm, however, does not consider temporal information of the Electroencephalogram (EEG) signals even though EEG signals are naturally non-stationary. In order to circumvent the limitation, in this paper, we propose a novel method, Time-Dependent Common Spatial Patterns (TDCSP) to classify multi-class motor imagery tasks. We optimize CSP filters in multiple local time ranges of EEG signals individually based on statistical analysis to effectively reflect changes of discriminative spatial distributions over time. We evaluated the proposed method by experiments on BCI Competition IV dataset 2-a, which resulted in high performance outperforming the previous methods in the literature.

Original languageEnglish
Title of host publication1st Asian Conference on Pattern Recognition, ACPR 2011
Pages643-646
Number of pages4
DOIs
Publication statusPublished - 2011 Dec 1
Event1st Asian Conference on Pattern Recognition, ACPR 2011 - Beijing, China
Duration: 2011 Nov 282011 Nov 28

Other

Other1st Asian Conference on Pattern Recognition, ACPR 2011
CountryChina
CityBeijing
Period11/11/2811/11/28

Fingerprint

Electroencephalography
Brain computer interface
Spatial distribution
Synchronization
Statistical methods
Experiments

Keywords

  • Brain-Computer Interface (BCI)
  • Common Spatial Patterns (CSP)
  • electroencephalogram (EEG)
  • motor imagery
  • Time-Dependent Common Spatial Patterns (TDCSP)

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition

Cite this

Kam, T. E., & Lee, S. W. (2011). Time-dependent common spatial patterns optimization for EEG signal classification. In 1st Asian Conference on Pattern Recognition, ACPR 2011 (pp. 643-646). [6166621] https://doi.org/10.1109/ACPR.2011.6166621

Time-dependent common spatial patterns optimization for EEG signal classification. / Kam, Tae Eui; Lee, Seong Whan.

1st Asian Conference on Pattern Recognition, ACPR 2011. 2011. p. 643-646 6166621.

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

Kam, TE & Lee, SW 2011, Time-dependent common spatial patterns optimization for EEG signal classification. in 1st Asian Conference on Pattern Recognition, ACPR 2011., 6166621, pp. 643-646, 1st Asian Conference on Pattern Recognition, ACPR 2011, Beijing, China, 11/11/28. https://doi.org/10.1109/ACPR.2011.6166621
Kam TE, Lee SW. Time-dependent common spatial patterns optimization for EEG signal classification. In 1st Asian Conference on Pattern Recognition, ACPR 2011. 2011. p. 643-646. 6166621 https://doi.org/10.1109/ACPR.2011.6166621
Kam, Tae Eui ; Lee, Seong Whan. / Time-dependent common spatial patterns optimization for EEG signal classification. 1st Asian Conference on Pattern Recognition, ACPR 2011. 2011. pp. 643-646
@inproceedings{a0cc2d5f16184ba28256dfe6d69998d6,
title = "Time-dependent common spatial patterns optimization for EEG signal classification",
abstract = "Recognizing Event-Related Desynchronization or Synchronization (ERD/ERS) patterns generated by motor imagery tasks is an important process in Brain-computer interfaces (BCI). One of the most well-known algorithms to extract the discriminative patterns is Common Spatial Patterns (CSP). It finds an optimal spatial filter considering the spatial distribution of the ERD/ERS patterns. The CSP algorithm, however, does not consider temporal information of the Electroencephalogram (EEG) signals even though EEG signals are naturally non-stationary. In order to circumvent the limitation, in this paper, we propose a novel method, Time-Dependent Common Spatial Patterns (TDCSP) to classify multi-class motor imagery tasks. We optimize CSP filters in multiple local time ranges of EEG signals individually based on statistical analysis to effectively reflect changes of discriminative spatial distributions over time. We evaluated the proposed method by experiments on BCI Competition IV dataset 2-a, which resulted in high performance outperforming the previous methods in the literature.",
keywords = "Brain-Computer Interface (BCI), Common Spatial Patterns (CSP), electroencephalogram (EEG), motor imagery, Time-Dependent Common Spatial Patterns (TDCSP)",
author = "Kam, {Tae Eui} and Lee, {Seong Whan}",
year = "2011",
month = "12",
day = "1",
doi = "10.1109/ACPR.2011.6166621",
language = "English",
isbn = "9781457701221",
pages = "643--646",
booktitle = "1st Asian Conference on Pattern Recognition, ACPR 2011",

}

TY - GEN

T1 - Time-dependent common spatial patterns optimization for EEG signal classification

AU - Kam, Tae Eui

AU - Lee, Seong Whan

PY - 2011/12/1

Y1 - 2011/12/1

N2 - Recognizing Event-Related Desynchronization or Synchronization (ERD/ERS) patterns generated by motor imagery tasks is an important process in Brain-computer interfaces (BCI). One of the most well-known algorithms to extract the discriminative patterns is Common Spatial Patterns (CSP). It finds an optimal spatial filter considering the spatial distribution of the ERD/ERS patterns. The CSP algorithm, however, does not consider temporal information of the Electroencephalogram (EEG) signals even though EEG signals are naturally non-stationary. In order to circumvent the limitation, in this paper, we propose a novel method, Time-Dependent Common Spatial Patterns (TDCSP) to classify multi-class motor imagery tasks. We optimize CSP filters in multiple local time ranges of EEG signals individually based on statistical analysis to effectively reflect changes of discriminative spatial distributions over time. We evaluated the proposed method by experiments on BCI Competition IV dataset 2-a, which resulted in high performance outperforming the previous methods in the literature.

AB - Recognizing Event-Related Desynchronization or Synchronization (ERD/ERS) patterns generated by motor imagery tasks is an important process in Brain-computer interfaces (BCI). One of the most well-known algorithms to extract the discriminative patterns is Common Spatial Patterns (CSP). It finds an optimal spatial filter considering the spatial distribution of the ERD/ERS patterns. The CSP algorithm, however, does not consider temporal information of the Electroencephalogram (EEG) signals even though EEG signals are naturally non-stationary. In order to circumvent the limitation, in this paper, we propose a novel method, Time-Dependent Common Spatial Patterns (TDCSP) to classify multi-class motor imagery tasks. We optimize CSP filters in multiple local time ranges of EEG signals individually based on statistical analysis to effectively reflect changes of discriminative spatial distributions over time. We evaluated the proposed method by experiments on BCI Competition IV dataset 2-a, which resulted in high performance outperforming the previous methods in the literature.

KW - Brain-Computer Interface (BCI)

KW - Common Spatial Patterns (CSP)

KW - electroencephalogram (EEG)

KW - motor imagery

KW - Time-Dependent Common Spatial Patterns (TDCSP)

UR - http://www.scopus.com/inward/record.url?scp=84862883608&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862883608&partnerID=8YFLogxK

U2 - 10.1109/ACPR.2011.6166621

DO - 10.1109/ACPR.2011.6166621

M3 - Conference contribution

AN - SCOPUS:84862883608

SN - 9781457701221

SP - 643

EP - 646

BT - 1st Asian Conference on Pattern Recognition, ACPR 2011

ER -