Non-homogeneous spatial filter optimization for ElectroEncephaloGram (EEG)-based motor imagery classification

Research output: Contribution to journalArticle

41 Citations (Scopus)

Abstract

Neuronal power attenuation or enhancement in specific frequency bands over the sensorimotor cortex, called Event-Related Desynchronization (ERD) or Event-Related Synchronization (ERS), respectively, is a major phenomenon in brain activities involved in imaginary movement of body parts. However, it is known that the nature of motor imagery-related electroencephalogram (EEG) signals is non-stationary and highly variable over time and frequency. In this paper, we propose a novel method of finding a discriminative time- and frequency-dependent spatial filter, which we call 'non-homogeneous filter.' We adaptively select bases of spatial filters over time and frequency. By taking both temporal and spectral features of EEGs in finding a spatial filter into account it is beneficial to be able to consider non-stationarity of EEG signals. In order to consider changes of ERD/ERS patterns over the time-frequency domain, we devise a spectrally and temporally weighted classification method via statistical analysis. Our experimental results on the BCI Competition IV dataset II-a and BCI Competition II dataset IV clearly presented the effectiveness of the proposed method outperforming other competing methods in the literature.

Original languageEnglish
Pages (from-to)58-68
Number of pages11
JournalNeurocomputing
Volume108
DOIs
Publication statusPublished - 2013 May 2

Fingerprint

Imagery (Psychotherapy)
Electroencephalography
Synchronization
Frequency bands
Brain
Statistical methods
Human Body
Datasets

Keywords

  • Brain-Computer Interface (BCI)
  • Electroencephalogram (EEG)
  • Motor imagery classification
  • Spatial filter optimization

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Cognitive Neuroscience

Cite this

Non-homogeneous spatial filter optimization for ElectroEncephaloGram (EEG)-based motor imagery classification. / Kam, Tae Eui; Suk, Heung-Il; Lee, Seong Whan.

In: Neurocomputing, Vol. 108, 02.05.2013, p. 58-68.

Research output: Contribution to journalArticle

@article{eb6fd56de88e4cf2890039a49262e5d7,
title = "Non-homogeneous spatial filter optimization for ElectroEncephaloGram (EEG)-based motor imagery classification",
abstract = "Neuronal power attenuation or enhancement in specific frequency bands over the sensorimotor cortex, called Event-Related Desynchronization (ERD) or Event-Related Synchronization (ERS), respectively, is a major phenomenon in brain activities involved in imaginary movement of body parts. However, it is known that the nature of motor imagery-related electroencephalogram (EEG) signals is non-stationary and highly variable over time and frequency. In this paper, we propose a novel method of finding a discriminative time- and frequency-dependent spatial filter, which we call 'non-homogeneous filter.' We adaptively select bases of spatial filters over time and frequency. By taking both temporal and spectral features of EEGs in finding a spatial filter into account it is beneficial to be able to consider non-stationarity of EEG signals. In order to consider changes of ERD/ERS patterns over the time-frequency domain, we devise a spectrally and temporally weighted classification method via statistical analysis. Our experimental results on the BCI Competition IV dataset II-a and BCI Competition II dataset IV clearly presented the effectiveness of the proposed method outperforming other competing methods in the literature.",
keywords = "Brain-Computer Interface (BCI), Electroencephalogram (EEG), Motor imagery classification, Spatial filter optimization",
author = "Kam, {Tae Eui} and Heung-Il Suk and Lee, {Seong Whan}",
year = "2013",
month = "5",
day = "2",
doi = "10.1016/j.neucom.2012.12.002",
language = "English",
volume = "108",
pages = "58--68",
journal = "Neurocomputing",
issn = "0925-2312",
publisher = "Elsevier",

}

TY - JOUR

T1 - Non-homogeneous spatial filter optimization for ElectroEncephaloGram (EEG)-based motor imagery classification

AU - Kam, Tae Eui

AU - Suk, Heung-Il

AU - Lee, Seong Whan

PY - 2013/5/2

Y1 - 2013/5/2

N2 - Neuronal power attenuation or enhancement in specific frequency bands over the sensorimotor cortex, called Event-Related Desynchronization (ERD) or Event-Related Synchronization (ERS), respectively, is a major phenomenon in brain activities involved in imaginary movement of body parts. However, it is known that the nature of motor imagery-related electroencephalogram (EEG) signals is non-stationary and highly variable over time and frequency. In this paper, we propose a novel method of finding a discriminative time- and frequency-dependent spatial filter, which we call 'non-homogeneous filter.' We adaptively select bases of spatial filters over time and frequency. By taking both temporal and spectral features of EEGs in finding a spatial filter into account it is beneficial to be able to consider non-stationarity of EEG signals. In order to consider changes of ERD/ERS patterns over the time-frequency domain, we devise a spectrally and temporally weighted classification method via statistical analysis. Our experimental results on the BCI Competition IV dataset II-a and BCI Competition II dataset IV clearly presented the effectiveness of the proposed method outperforming other competing methods in the literature.

AB - Neuronal power attenuation or enhancement in specific frequency bands over the sensorimotor cortex, called Event-Related Desynchronization (ERD) or Event-Related Synchronization (ERS), respectively, is a major phenomenon in brain activities involved in imaginary movement of body parts. However, it is known that the nature of motor imagery-related electroencephalogram (EEG) signals is non-stationary and highly variable over time and frequency. In this paper, we propose a novel method of finding a discriminative time- and frequency-dependent spatial filter, which we call 'non-homogeneous filter.' We adaptively select bases of spatial filters over time and frequency. By taking both temporal and spectral features of EEGs in finding a spatial filter into account it is beneficial to be able to consider non-stationarity of EEG signals. In order to consider changes of ERD/ERS patterns over the time-frequency domain, we devise a spectrally and temporally weighted classification method via statistical analysis. Our experimental results on the BCI Competition IV dataset II-a and BCI Competition II dataset IV clearly presented the effectiveness of the proposed method outperforming other competing methods in the literature.

KW - Brain-Computer Interface (BCI)

KW - Electroencephalogram (EEG)

KW - Motor imagery classification

KW - Spatial filter optimization

UR - http://www.scopus.com/inward/record.url?scp=84875095765&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84875095765&partnerID=8YFLogxK

U2 - 10.1016/j.neucom.2012.12.002

DO - 10.1016/j.neucom.2012.12.002

M3 - Article

AN - SCOPUS:84875095765

VL - 108

SP - 58

EP - 68

JO - Neurocomputing

JF - Neurocomputing

SN - 0925-2312

ER -