Towards adaptive classification for BCI

Pradeep Shenoy, Matthias Krauledat, Benjamin Blankertz, Rajesh P N Rao, Klaus Muller

Research output: Contribution to journalArticle

291 Citations (Scopus)

Abstract

Non-stationarities are ubiquitous in EEG signals. They are especially apparent in the use of EEG-based brain-computer interfaces (BCIs): (a) in the differences between the initial calibration measurement and the online operation of a BCI, or (b) caused by changes in the subject's brain processes during an experiment (e.g. due to fatigue, change of task involvement, etc). In this paper, we quantify for the first time such systematic evidence of statistical differences in data recorded during offline and online sessions. Furthermore, we propose novel techniques of investigating and visualizing data distributions, which are particularly useful for the analysis of (non-)stationarities. Our study shows that the brain signals used for control can change substantially from the offline calibration sessions to online control, and also within a single session. In addition to this general characterization of the signals, we propose several adaptive classification schemes and study their performance on data recorded during online experiments. An encouraging result of our study is that surprisingly simple adaptive methods in combination with an offline feature selection scheme can significantly increase BCI performance.

Original languageEnglish
JournalJournal of Neural Engineering
Volume3
Issue number1
DOIs
Publication statusPublished - 2006 Mar 1
Externally publishedYes

Fingerprint

Brain-Computer Interfaces
Brain computer interface
Electroencephalography
Calibration
Brain
Fatigue
Feature extraction
Experiments
Fatigue of materials

ASJC Scopus subject areas

  • Biotechnology
  • Bioengineering
  • Neuroscience (miscellaneous)

Cite this

Shenoy, P., Krauledat, M., Blankertz, B., Rao, R. P. N., & Muller, K. (2006). Towards adaptive classification for BCI. Journal of Neural Engineering, 3(1). https://doi.org/10.1088/1741-2560/3/1/R02

Towards adaptive classification for BCI. / Shenoy, Pradeep; Krauledat, Matthias; Blankertz, Benjamin; Rao, Rajesh P N; Muller, Klaus.

In: Journal of Neural Engineering, Vol. 3, No. 1, 01.03.2006.

Research output: Contribution to journalArticle

Shenoy, P, Krauledat, M, Blankertz, B, Rao, RPN & Muller, K 2006, 'Towards adaptive classification for BCI', Journal of Neural Engineering, vol. 3, no. 1. https://doi.org/10.1088/1741-2560/3/1/R02
Shenoy P, Krauledat M, Blankertz B, Rao RPN, Muller K. Towards adaptive classification for BCI. Journal of Neural Engineering. 2006 Mar 1;3(1). https://doi.org/10.1088/1741-2560/3/1/R02
Shenoy, Pradeep ; Krauledat, Matthias ; Blankertz, Benjamin ; Rao, Rajesh P N ; Muller, Klaus. / Towards adaptive classification for BCI. In: Journal of Neural Engineering. 2006 ; Vol. 3, No. 1.
@article{14e219fafbac47aa81c8cc6d2cb462a2,
title = "Towards adaptive classification for BCI",
abstract = "Non-stationarities are ubiquitous in EEG signals. They are especially apparent in the use of EEG-based brain-computer interfaces (BCIs): (a) in the differences between the initial calibration measurement and the online operation of a BCI, or (b) caused by changes in the subject's brain processes during an experiment (e.g. due to fatigue, change of task involvement, etc). In this paper, we quantify for the first time such systematic evidence of statistical differences in data recorded during offline and online sessions. Furthermore, we propose novel techniques of investigating and visualizing data distributions, which are particularly useful for the analysis of (non-)stationarities. Our study shows that the brain signals used for control can change substantially from the offline calibration sessions to online control, and also within a single session. In addition to this general characterization of the signals, we propose several adaptive classification schemes and study their performance on data recorded during online experiments. An encouraging result of our study is that surprisingly simple adaptive methods in combination with an offline feature selection scheme can significantly increase BCI performance.",
author = "Pradeep Shenoy and Matthias Krauledat and Benjamin Blankertz and Rao, {Rajesh P N} and Klaus Muller",
year = "2006",
month = "3",
day = "1",
doi = "10.1088/1741-2560/3/1/R02",
language = "English",
volume = "3",
journal = "Journal of Neural Engineering",
issn = "1741-2560",
publisher = "IOP Publishing Ltd.",
number = "1",

}

TY - JOUR

T1 - Towards adaptive classification for BCI

AU - Shenoy, Pradeep

AU - Krauledat, Matthias

AU - Blankertz, Benjamin

AU - Rao, Rajesh P N

AU - Muller, Klaus

PY - 2006/3/1

Y1 - 2006/3/1

N2 - Non-stationarities are ubiquitous in EEG signals. They are especially apparent in the use of EEG-based brain-computer interfaces (BCIs): (a) in the differences between the initial calibration measurement and the online operation of a BCI, or (b) caused by changes in the subject's brain processes during an experiment (e.g. due to fatigue, change of task involvement, etc). In this paper, we quantify for the first time such systematic evidence of statistical differences in data recorded during offline and online sessions. Furthermore, we propose novel techniques of investigating and visualizing data distributions, which are particularly useful for the analysis of (non-)stationarities. Our study shows that the brain signals used for control can change substantially from the offline calibration sessions to online control, and also within a single session. In addition to this general characterization of the signals, we propose several adaptive classification schemes and study their performance on data recorded during online experiments. An encouraging result of our study is that surprisingly simple adaptive methods in combination with an offline feature selection scheme can significantly increase BCI performance.

AB - Non-stationarities are ubiquitous in EEG signals. They are especially apparent in the use of EEG-based brain-computer interfaces (BCIs): (a) in the differences between the initial calibration measurement and the online operation of a BCI, or (b) caused by changes in the subject's brain processes during an experiment (e.g. due to fatigue, change of task involvement, etc). In this paper, we quantify for the first time such systematic evidence of statistical differences in data recorded during offline and online sessions. Furthermore, we propose novel techniques of investigating and visualizing data distributions, which are particularly useful for the analysis of (non-)stationarities. Our study shows that the brain signals used for control can change substantially from the offline calibration sessions to online control, and also within a single session. In addition to this general characterization of the signals, we propose several adaptive classification schemes and study their performance on data recorded during online experiments. An encouraging result of our study is that surprisingly simple adaptive methods in combination with an offline feature selection scheme can significantly increase BCI performance.

UR - http://www.scopus.com/inward/record.url?scp=33644692705&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33644692705&partnerID=8YFLogxK

U2 - 10.1088/1741-2560/3/1/R02

DO - 10.1088/1741-2560/3/1/R02

M3 - Article

C2 - 16510936

AN - SCOPUS:33644692705

VL - 3

JO - Journal of Neural Engineering

JF - Journal of Neural Engineering

SN - 1741-2560

IS - 1

ER -