A Combination of Pre- and Postprocessing Techniques to Enhance Self-Paced BCIs

Raheleh Mohammadi, Ali Mahloojifar, DH Coyle

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Mental task onset detection from the continuous electroencephalogram (EEG) in real time is a critical issue in self-paced brain computer interface (BCI) design. The paper shows that self-paced BCI performance can be significantly improved by combining a range of simple techniques including (1) constant-Q filters with varying bandwidth size depending on the center frequency, instead of constant bandwidth filters for frequency decomposition of the EEG signal in the 6 to 36 Hz band; (2) subjectspecific postprocessing parameter optimization consisting of dwell time and threshold, and (3) debiasing before postprocessing by readjusting the classification output based on the current and previous brain states, to reduce the number of false detections. This debiasing block is shown to be optimal when activated only in special cases which are predetermined during the training phase. Analysis of the data recorded from seven subjects executing foot movement shows a statistically significant 10% (P <0.05) average improvement in true positive rate (TPR) and a 1% reduction in false positive rate (FPR) detections compared with previous work on the same data.
LanguageEnglish
Pages1-10
JournalAdvances in Human-Computer Interaction
Volume2012
Issue number185320
DOIs
Publication statusPublished - 2012

Fingerprint

Brain computer interface
Electroencephalography
Bandwidth
Brain
Decomposition

Cite this

Mohammadi, Raheleh ; Mahloojifar, Ali ; Coyle, DH. / A Combination of Pre- and Postprocessing Techniques to Enhance Self-Paced BCIs. In: Advances in Human-Computer Interaction. 2012 ; Vol. 2012, No. 185320. pp. 1-10.
@article{fda6408c39c7420fb919beb3bae83e64,
title = "A Combination of Pre- and Postprocessing Techniques to Enhance Self-Paced BCIs",
abstract = "Mental task onset detection from the continuous electroencephalogram (EEG) in real time is a critical issue in self-paced brain computer interface (BCI) design. The paper shows that self-paced BCI performance can be significantly improved by combining a range of simple techniques including (1) constant-Q filters with varying bandwidth size depending on the center frequency, instead of constant bandwidth filters for frequency decomposition of the EEG signal in the 6 to 36 Hz band; (2) subjectspecific postprocessing parameter optimization consisting of dwell time and threshold, and (3) debiasing before postprocessing by readjusting the classification output based on the current and previous brain states, to reduce the number of false detections. This debiasing block is shown to be optimal when activated only in special cases which are predetermined during the training phase. Analysis of the data recorded from seven subjects executing foot movement shows a statistically significant 10{\%} (P <0.05) average improvement in true positive rate (TPR) and a 1{\%} reduction in false positive rate (FPR) detections compared with previous work on the same data.",
author = "Raheleh Mohammadi and Ali Mahloojifar and DH Coyle",
year = "2012",
doi = "10.1155/2012/185320",
language = "English",
volume = "2012",
pages = "1--10",
journal = "Advances in Human-Computer Interaction",
issn = "1687-5893",
number = "185320",

}

A Combination of Pre- and Postprocessing Techniques to Enhance Self-Paced BCIs. / Mohammadi, Raheleh; Mahloojifar, Ali; Coyle, DH.

In: Advances in Human-Computer Interaction, Vol. 2012, No. 185320, 2012, p. 1-10.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A Combination of Pre- and Postprocessing Techniques to Enhance Self-Paced BCIs

AU - Mohammadi, Raheleh

AU - Mahloojifar, Ali

AU - Coyle, DH

PY - 2012

Y1 - 2012

N2 - Mental task onset detection from the continuous electroencephalogram (EEG) in real time is a critical issue in self-paced brain computer interface (BCI) design. The paper shows that self-paced BCI performance can be significantly improved by combining a range of simple techniques including (1) constant-Q filters with varying bandwidth size depending on the center frequency, instead of constant bandwidth filters for frequency decomposition of the EEG signal in the 6 to 36 Hz band; (2) subjectspecific postprocessing parameter optimization consisting of dwell time and threshold, and (3) debiasing before postprocessing by readjusting the classification output based on the current and previous brain states, to reduce the number of false detections. This debiasing block is shown to be optimal when activated only in special cases which are predetermined during the training phase. Analysis of the data recorded from seven subjects executing foot movement shows a statistically significant 10% (P <0.05) average improvement in true positive rate (TPR) and a 1% reduction in false positive rate (FPR) detections compared with previous work on the same data.

AB - Mental task onset detection from the continuous electroencephalogram (EEG) in real time is a critical issue in self-paced brain computer interface (BCI) design. The paper shows that self-paced BCI performance can be significantly improved by combining a range of simple techniques including (1) constant-Q filters with varying bandwidth size depending on the center frequency, instead of constant bandwidth filters for frequency decomposition of the EEG signal in the 6 to 36 Hz band; (2) subjectspecific postprocessing parameter optimization consisting of dwell time and threshold, and (3) debiasing before postprocessing by readjusting the classification output based on the current and previous brain states, to reduce the number of false detections. This debiasing block is shown to be optimal when activated only in special cases which are predetermined during the training phase. Analysis of the data recorded from seven subjects executing foot movement shows a statistically significant 10% (P <0.05) average improvement in true positive rate (TPR) and a 1% reduction in false positive rate (FPR) detections compared with previous work on the same data.

U2 - 10.1155/2012/185320

DO - 10.1155/2012/185320

M3 - Article

VL - 2012

SP - 1

EP - 10

JO - Advances in Human-Computer Interaction

T2 - Advances in Human-Computer Interaction

JF - Advances in Human-Computer Interaction

SN - 1687-5893

IS - 185320

ER -