The Berlin brain-computer interface: Machine learning based detection of user specific brain states

Benjamin Blankertz, Guido Dornhege, Steven Lemm, Matthias Krauledat, Gabriel Curio, Klaus Muller

Research output: Contribution to journalArticle

77 Citations (Scopus)

Abstract

We outline the Berlin Brain-Computer Interface (BBCI), a system which enables us to translate brain signals from movements or movement intentions into control commands. The main contribution of the BBCI, which is a non-invasive EEG-based BCI system, is the use of advanced machine learning techniques that allow to adapt to the specific brain signatures of each user with literally no training. In BBCI a calibration session of about 20min is necessary to provide a data basis from which the individualized brain signatures are inferred. This is very much in contrast to conventional BCI approaches that rely on operand conditioning and need extensive subject training of the order 50-100 hours. Our machine learning concept thus allows to achieve high quality feedback already after the very first session. This work reviews a broad range of investigations and experiments that have been performed within the BBCI project. In addition to these general paradigmatic BCI results, this work provides a condensed outline of the underlying machine learning and signal processing techniques that make the BBCI succeed. In the first experimental paradigm we analyze the predictability of limb movement long before the actual movement takes place using only the movement intention measured from the pre-movement (readiness) EEG potentials. The experiments include both off-line studies and an online feedback paradigm. The limits with respect to the spatial resolution of the somatotopy are explored by contrasting brain patterns of movements of left vs. right hand rsp. foot. In a second complementary paradigm voluntary modulations of sensorimotor rhythms caused by motor imagery (left hand vs. right hand vs. foot) are translated into a continuous feedback signal. Here we report results of a recent feedback study with 6 healthy subjects with no or very little experience with BCI control: half of the subjects achieved an information transfer rate above 35 bits per minute (bpm). Furthermore one subject used the BBCI to operate a mental typewriter in free spelling mode. The overall spelling speed was 4.5-8 letters per minute including the time needed for the correction errors.

Original languageEnglish
Pages (from-to)581-607
Number of pages27
JournalJournal of Universal Computer Science
Volume12
Issue number6
Publication statusPublished - 2006 Aug 11
Externally publishedYes

Fingerprint

Brain computer interface
Learning systems
Brain
Machine Learning
Feedback
Electroencephalography
Typewriters
Paradigm
Bioelectric potentials
Error correction
Signature
Signal processing
Experiments
Modulation
Calibration
Information Transfer
Predictability
Error Correction
Movement
Conditioning

Keywords

  • Brain-Computer Interface
  • Classification
  • Common Spatial Patterns
  • EEg
  • ERD
  • Event-Related Desynchronization
  • Feedback
  • Information Transfer Rate
  • Machine Learning
  • Readiness Potential
  • RP
  • Single-Trial Analysis

ASJC Scopus subject areas

  • Computer Science(all)

Cite this

Blankertz, B., Dornhege, G., Lemm, S., Krauledat, M., Curio, G., & Muller, K. (2006). The Berlin brain-computer interface: Machine learning based detection of user specific brain states. Journal of Universal Computer Science, 12(6), 581-607.

The Berlin brain-computer interface : Machine learning based detection of user specific brain states. / Blankertz, Benjamin; Dornhege, Guido; Lemm, Steven; Krauledat, Matthias; Curio, Gabriel; Muller, Klaus.

In: Journal of Universal Computer Science, Vol. 12, No. 6, 11.08.2006, p. 581-607.

Research output: Contribution to journalArticle

Blankertz, B, Dornhege, G, Lemm, S, Krauledat, M, Curio, G & Muller, K 2006, 'The Berlin brain-computer interface: Machine learning based detection of user specific brain states', Journal of Universal Computer Science, vol. 12, no. 6, pp. 581-607.
Blankertz B, Dornhege G, Lemm S, Krauledat M, Curio G, Muller K. The Berlin brain-computer interface: Machine learning based detection of user specific brain states. Journal of Universal Computer Science. 2006 Aug 11;12(6):581-607.
Blankertz, Benjamin ; Dornhege, Guido ; Lemm, Steven ; Krauledat, Matthias ; Curio, Gabriel ; Muller, Klaus. / The Berlin brain-computer interface : Machine learning based detection of user specific brain states. In: Journal of Universal Computer Science. 2006 ; Vol. 12, No. 6. pp. 581-607.
@article{fa0b258a1dd447a7b0b2a03e09382b88,
title = "The Berlin brain-computer interface: Machine learning based detection of user specific brain states",
abstract = "We outline the Berlin Brain-Computer Interface (BBCI), a system which enables us to translate brain signals from movements or movement intentions into control commands. The main contribution of the BBCI, which is a non-invasive EEG-based BCI system, is the use of advanced machine learning techniques that allow to adapt to the specific brain signatures of each user with literally no training. In BBCI a calibration session of about 20min is necessary to provide a data basis from which the individualized brain signatures are inferred. This is very much in contrast to conventional BCI approaches that rely on operand conditioning and need extensive subject training of the order 50-100 hours. Our machine learning concept thus allows to achieve high quality feedback already after the very first session. This work reviews a broad range of investigations and experiments that have been performed within the BBCI project. In addition to these general paradigmatic BCI results, this work provides a condensed outline of the underlying machine learning and signal processing techniques that make the BBCI succeed. In the first experimental paradigm we analyze the predictability of limb movement long before the actual movement takes place using only the movement intention measured from the pre-movement (readiness) EEG potentials. The experiments include both off-line studies and an online feedback paradigm. The limits with respect to the spatial resolution of the somatotopy are explored by contrasting brain patterns of movements of left vs. right hand rsp. foot. In a second complementary paradigm voluntary modulations of sensorimotor rhythms caused by motor imagery (left hand vs. right hand vs. foot) are translated into a continuous feedback signal. Here we report results of a recent feedback study with 6 healthy subjects with no or very little experience with BCI control: half of the subjects achieved an information transfer rate above 35 bits per minute (bpm). Furthermore one subject used the BBCI to operate a mental typewriter in free spelling mode. The overall spelling speed was 4.5-8 letters per minute including the time needed for the correction errors.",
keywords = "Brain-Computer Interface, Classification, Common Spatial Patterns, EEg, ERD, Event-Related Desynchronization, Feedback, Information Transfer Rate, Machine Learning, Readiness Potential, RP, Single-Trial Analysis",
author = "Benjamin Blankertz and Guido Dornhege and Steven Lemm and Matthias Krauledat and Gabriel Curio and Klaus Muller",
year = "2006",
month = "8",
day = "11",
language = "English",
volume = "12",
pages = "581--607",
journal = "Journal of Universal Computer Science",
issn = "0948-6968",
publisher = "Springer Verlag",
number = "6",

}

TY - JOUR

T1 - The Berlin brain-computer interface

T2 - Machine learning based detection of user specific brain states

AU - Blankertz, Benjamin

AU - Dornhege, Guido

AU - Lemm, Steven

AU - Krauledat, Matthias

AU - Curio, Gabriel

AU - Muller, Klaus

PY - 2006/8/11

Y1 - 2006/8/11

N2 - We outline the Berlin Brain-Computer Interface (BBCI), a system which enables us to translate brain signals from movements or movement intentions into control commands. The main contribution of the BBCI, which is a non-invasive EEG-based BCI system, is the use of advanced machine learning techniques that allow to adapt to the specific brain signatures of each user with literally no training. In BBCI a calibration session of about 20min is necessary to provide a data basis from which the individualized brain signatures are inferred. This is very much in contrast to conventional BCI approaches that rely on operand conditioning and need extensive subject training of the order 50-100 hours. Our machine learning concept thus allows to achieve high quality feedback already after the very first session. This work reviews a broad range of investigations and experiments that have been performed within the BBCI project. In addition to these general paradigmatic BCI results, this work provides a condensed outline of the underlying machine learning and signal processing techniques that make the BBCI succeed. In the first experimental paradigm we analyze the predictability of limb movement long before the actual movement takes place using only the movement intention measured from the pre-movement (readiness) EEG potentials. The experiments include both off-line studies and an online feedback paradigm. The limits with respect to the spatial resolution of the somatotopy are explored by contrasting brain patterns of movements of left vs. right hand rsp. foot. In a second complementary paradigm voluntary modulations of sensorimotor rhythms caused by motor imagery (left hand vs. right hand vs. foot) are translated into a continuous feedback signal. Here we report results of a recent feedback study with 6 healthy subjects with no or very little experience with BCI control: half of the subjects achieved an information transfer rate above 35 bits per minute (bpm). Furthermore one subject used the BBCI to operate a mental typewriter in free spelling mode. The overall spelling speed was 4.5-8 letters per minute including the time needed for the correction errors.

AB - We outline the Berlin Brain-Computer Interface (BBCI), a system which enables us to translate brain signals from movements or movement intentions into control commands. The main contribution of the BBCI, which is a non-invasive EEG-based BCI system, is the use of advanced machine learning techniques that allow to adapt to the specific brain signatures of each user with literally no training. In BBCI a calibration session of about 20min is necessary to provide a data basis from which the individualized brain signatures are inferred. This is very much in contrast to conventional BCI approaches that rely on operand conditioning and need extensive subject training of the order 50-100 hours. Our machine learning concept thus allows to achieve high quality feedback already after the very first session. This work reviews a broad range of investigations and experiments that have been performed within the BBCI project. In addition to these general paradigmatic BCI results, this work provides a condensed outline of the underlying machine learning and signal processing techniques that make the BBCI succeed. In the first experimental paradigm we analyze the predictability of limb movement long before the actual movement takes place using only the movement intention measured from the pre-movement (readiness) EEG potentials. The experiments include both off-line studies and an online feedback paradigm. The limits with respect to the spatial resolution of the somatotopy are explored by contrasting brain patterns of movements of left vs. right hand rsp. foot. In a second complementary paradigm voluntary modulations of sensorimotor rhythms caused by motor imagery (left hand vs. right hand vs. foot) are translated into a continuous feedback signal. Here we report results of a recent feedback study with 6 healthy subjects with no or very little experience with BCI control: half of the subjects achieved an information transfer rate above 35 bits per minute (bpm). Furthermore one subject used the BBCI to operate a mental typewriter in free spelling mode. The overall spelling speed was 4.5-8 letters per minute including the time needed for the correction errors.

KW - Brain-Computer Interface

KW - Classification

KW - Common Spatial Patterns

KW - EEg

KW - ERD

KW - Event-Related Desynchronization

KW - Feedback

KW - Information Transfer Rate

KW - Machine Learning

KW - Readiness Potential

KW - RP

KW - Single-Trial Analysis

UR - http://www.scopus.com/inward/record.url?scp=33746841002&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33746841002&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:33746841002

VL - 12

SP - 581

EP - 607

JO - Journal of Universal Computer Science

JF - Journal of Universal Computer Science

SN - 0948-6968

IS - 6

ER -