A subject-independent brain-computer interface based on smoothed, second-order baselining

B. Reuderink, Jason Farquhar, Mannes Poel, Antinus Nijholt

  • 9 Citations

Abstract

A brain-computer interface (BCI) enables direct communication from the brain to devices, bypassing the traditional pathway of peripheral nerves and muscles. Traditional approaches to BCIs require the user to train for weeks or even months to learn to control the BCI. In contrast, BCIs based on machine learning only require a calibration session of less than an hour before the system can be used, since the machine adapts to the user’s existing brain signals. However, this calibration session has to be repeated before each use of the BCI due to inter-session variability, which makes using a BCI still a time-consuming and an error-prone enterprise. In this work, we present a second-order baselining procedure that reduces these variations, and enables the creation of a BCI that can be applied to new subjects without such a calibration session. The method was validated with a motor-imagery classification task performed by 109 subjects. Results showed that our subjectindependent BCI without calibration performs as well as the popular common spatial patterns (CSP)-based BCI that does use a calibration session.
Original languageUndefined
Title of host publicationProceedings 33rd Annual IEEE Conference on Engineering in Medicine and Biology (EMBC)
Place of PublicationUSA
PublisherIEEE Engineering in Medicine and Biology Society
Pages4600-4604
Number of pages5
ISBN (Print)978-1-4244-4121-1
DOIs
StatePublished - 3 Sep 2011
Event33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2011 - Boston, United States

Publication series

Name
PublisherIEEE Engineering in Medicine & Biology Society

Conference

Conference33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2011
Abbreviated titleEMBC
CountryUnited States
CityBoston
Period30/08/113/09/11

Fingerprint

Brain computer interface
Calibration
Brain
Muscle
Learning systems
Communication
Industry

Keywords

  • METIS-281610
  • Machine Learning
  • Brain-Computer Interface
  • EWI-20919
  • IR-78787
  • Training
  • HMI-CI: Computational Intelligence
  • Calibration

Cite this

Reuderink, B., Farquhar, J., Poel, M., & Nijholt, A. (2011). A subject-independent brain-computer interface based on smoothed, second-order baselining. In Proceedings 33rd Annual IEEE Conference on Engineering in Medicine and Biology (EMBC) (pp. 4600-4604). USA: IEEE Engineering in Medicine and Biology Society. DOI: 10.1109/IEMBS.2011.6091139

Reuderink, B.; Farquhar, Jason; Poel, Mannes; Nijholt, Antinus / A subject-independent brain-computer interface based on smoothed, second-order baselining.

Proceedings 33rd Annual IEEE Conference on Engineering in Medicine and Biology (EMBC). USA : IEEE Engineering in Medicine and Biology Society, 2011. p. 4600-4604.

Research output: Scientific - peer-reviewConference contribution

@inbook{48727c34e1f943aa82927c0434cbb6ee,
title = "A subject-independent brain-computer interface based on smoothed, second-order baselining",
abstract = "A brain-computer interface (BCI) enables direct communication from the brain to devices, bypassing the traditional pathway of peripheral nerves and muscles. Traditional approaches to BCIs require the user to train for weeks or even months to learn to control the BCI. In contrast, BCIs based on machine learning only require a calibration session of less than an hour before the system can be used, since the machine adapts to the user’s existing brain signals. However, this calibration session has to be repeated before each use of the BCI due to inter-session variability, which makes using a BCI still a time-consuming and an error-prone enterprise. In this work, we present a second-order baselining procedure that reduces these variations, and enables the creation of a BCI that can be applied to new subjects without such a calibration session. The method was validated with a motor-imagery classification task performed by 109 subjects. Results showed that our subjectindependent BCI without calibration performs as well as the popular common spatial patterns (CSP)-based BCI that does use a calibration session.",
keywords = "METIS-281610, Machine Learning, Brain-Computer Interface, EWI-20919, IR-78787, Training, HMI-CI: Computational Intelligence, Calibration",
author = "B. Reuderink and Jason Farquhar and Mannes Poel and Antinus Nijholt",
note = "eemcs-eprint-20919",
year = "2011",
month = "9",
doi = "10.1109/IEMBS.2011.6091139",
isbn = "978-1-4244-4121-1",
publisher = "IEEE Engineering in Medicine and Biology Society",
pages = "4600--4604",
booktitle = "Proceedings 33rd Annual IEEE Conference on Engineering in Medicine and Biology (EMBC)",

}

Reuderink, B, Farquhar, J, Poel, M & Nijholt, A 2011, A subject-independent brain-computer interface based on smoothed, second-order baselining. in Proceedings 33rd Annual IEEE Conference on Engineering in Medicine and Biology (EMBC). IEEE Engineering in Medicine and Biology Society, USA, pp. 4600-4604, 33rd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2011, Boston, United States, 30-3 September. DOI: 10.1109/IEMBS.2011.6091139

A subject-independent brain-computer interface based on smoothed, second-order baselining. / Reuderink, B.; Farquhar, Jason; Poel, Mannes; Nijholt, Antinus.

Proceedings 33rd Annual IEEE Conference on Engineering in Medicine and Biology (EMBC). USA : IEEE Engineering in Medicine and Biology Society, 2011. p. 4600-4604.

Research output: Scientific - peer-reviewConference contribution

TY - CHAP

T1 - A subject-independent brain-computer interface based on smoothed, second-order baselining

AU - Reuderink,B.

AU - Farquhar,Jason

AU - Poel,Mannes

AU - Nijholt,Antinus

N1 - eemcs-eprint-20919

PY - 2011/9/3

Y1 - 2011/9/3

N2 - A brain-computer interface (BCI) enables direct communication from the brain to devices, bypassing the traditional pathway of peripheral nerves and muscles. Traditional approaches to BCIs require the user to train for weeks or even months to learn to control the BCI. In contrast, BCIs based on machine learning only require a calibration session of less than an hour before the system can be used, since the machine adapts to the user’s existing brain signals. However, this calibration session has to be repeated before each use of the BCI due to inter-session variability, which makes using a BCI still a time-consuming and an error-prone enterprise. In this work, we present a second-order baselining procedure that reduces these variations, and enables the creation of a BCI that can be applied to new subjects without such a calibration session. The method was validated with a motor-imagery classification task performed by 109 subjects. Results showed that our subjectindependent BCI without calibration performs as well as the popular common spatial patterns (CSP)-based BCI that does use a calibration session.

AB - A brain-computer interface (BCI) enables direct communication from the brain to devices, bypassing the traditional pathway of peripheral nerves and muscles. Traditional approaches to BCIs require the user to train for weeks or even months to learn to control the BCI. In contrast, BCIs based on machine learning only require a calibration session of less than an hour before the system can be used, since the machine adapts to the user’s existing brain signals. However, this calibration session has to be repeated before each use of the BCI due to inter-session variability, which makes using a BCI still a time-consuming and an error-prone enterprise. In this work, we present a second-order baselining procedure that reduces these variations, and enables the creation of a BCI that can be applied to new subjects without such a calibration session. The method was validated with a motor-imagery classification task performed by 109 subjects. Results showed that our subjectindependent BCI without calibration performs as well as the popular common spatial patterns (CSP)-based BCI that does use a calibration session.

KW - METIS-281610

KW - Machine Learning

KW - Brain-Computer Interface

KW - EWI-20919

KW - IR-78787

KW - Training

KW - HMI-CI: Computational Intelligence

KW - Calibration

U2 - 10.1109/IEMBS.2011.6091139

DO - 10.1109/IEMBS.2011.6091139

M3 - Conference contribution

SN - 978-1-4244-4121-1

SP - 4600

EP - 4604

BT - Proceedings 33rd Annual IEEE Conference on Engineering in Medicine and Biology (EMBC)

PB - IEEE Engineering in Medicine and Biology Society

ER -

Reuderink B, Farquhar J, Poel M, Nijholt A. A subject-independent brain-computer interface based on smoothed, second-order baselining. In Proceedings 33rd Annual IEEE Conference on Engineering in Medicine and Biology (EMBC). USA: IEEE Engineering in Medicine and Biology Society. 2011. p. 4600-4604. Available from, DOI: 10.1109/IEMBS.2011.6091139