Machine learning for real-time single-trial EEG-analysis: From brain–computer interfacing to mental state monitoring

Müller, Klaus Robert; Tangermann, Michael; Dornhege, Guido; Krauledat, Matthias; Curio, Gabriel; Blankertz, Benjamin

doi:10.1016/j.jneumeth.2007.09.022

Cited by 419 publications

(242 citation statements)

References 33 publications

Supporting

Mentioning

234

Contrasting

Unclassified

Order By: Relevance

“…Noninvasive BCIs use electroencephalographic (EEG) activity recorded from the scalp (Farwell and Donchin, 1988;Wolpaw et al, 1991;Sutter, 1992;McFarland et al, 1993McFarland et al, , 2008Pfurtscheller et al, 1993Pfurtscheller et al, , 2000McFarland, 1994, 2004;Birbaumer et al, 1999;Millán et al, 2004;Kübler et al, 2005;Blankertz et al, 2006;Vaughan et al, 2006;Müller et al, 2008). Although EEG-based BCIs support higher performance than often assumed, the acquisition of high levels of brain-based control typically requires extensive user training, and BCI performance can also be variable.…”

Section: Translating Microscale Neural Interface Technologies For CLImentioning

confidence: 99%

Advanced Neurotechnologies for Chronic Neural Interfaces: New Horizons and Clinical Opportunities

Kipke¹,

Shain²,

Buzsáki³

et al. 2008

J. Neurosci.

261

176

View full text Add to dashboard Cite

IntroductionTechnological advances in neural interfaces are providing increasingly more powerful "toolkits" of designs, materials, components, and integrated devices for establishing high-fidelity chronic neural interfaces. For a broad class of neuroscience studies, the primary requirements of these interfaces include recording and/or stimulating from a number of discretely sampled volumes at requisite spatial resolutions for specific periods of time. Translational and clinical applications present additional requirements for safety, usability, reliability, patient acceptance, and cost effectiveness. Innovative solutions result from the constructive tension between ever-increasing application requirements and incorporation of technological advances into usable devices. The purpose of this minireview is to present snapshots of the current state-of-the-art in chronic, microscale neural interfaces by highlighting several leading neuroscience applications and discussing their implications for next-generation interface devices.

show abstract

Section: Translating Microscale Neural Interface Technologies For CLImentioning

confidence: 99%

Advanced Neurotechnologies for Chronic Neural Interfaces: New Horizons and Clinical Opportunities

Kipke¹,

Shain²,

Buzsáki³

et al. 2008

J. Neurosci.

261

176

View full text Add to dashboard Cite

show abstract

“…This ability can be taken as a basis for Brain-Computer Interfaces (BCIs) which are devices that translate the intent of a subject, as derived from measuring brain signals, directly into control commands, e.g., for a computer application or a neuroprosthesis ; Allison et al (2007) ;Birbaumer et al (2006); Pfurtscheller et al (2005); Wolpaw et al (2002); Kübler et al (2001)). For alternative applications of BCI technology, see Kohlmorgen et al (2007); Müller et al (2008); Gerson et al (2006); Tangermann et al (2009). More specifically, SMR-based BCI operation has its neurophysiological grounds in the observation that macroscopic brain activity during resting wakefulness is comprised of distinct 'idle' rhythms located over distinct brain areas, e.g., the parieto-occipital α-rhythm (8-12 Hz), cf.…”

Section: Introductionmentioning

confidence: 99%

Neurophysiological predictor of SMR-based BCI performance

et al. 2010

Self Cite

View full text Add to dashboard Cite

Brain-Computer Interfaces (BCIs) allow a user to control a computer application by brain activity as measured, e.g., by electroencephalography (EEG).After about 30 years of BCI research, the success of control that is achieved by means of a BCI system still greatly varies between subjects. For about 20% of potential users the obtained accuracy does not reach the level criterion, meaning that BCI control is not accurate enough to control an application.The determination of factors that may serve to predict BCI performance, and the development of methods to quantify a predictor value from psychological and/or physiological data serves two purposes: a better understanding of the 'BCI-illiteracy phenomenon', and avoidance of a costly and eventually * Corresponding author. Berlin Institute of Technology, Machine Learning Laboratory, Sekr. FR6-9, Franklinstrasse 28/29, 10587 Berlin, Germany. Tel: +49 3031478624, Fax: +49 3031478622. E-mail address: dickhaus@cs.tu-berlin.de (Th. Dickhaus). Preprint submitted to NeuroImageMarch 8 which operates on modulations of sensory motor rhythms (SMRs).

show abstract

“…EEG is used to monitor ictal and inter-ictal activity in seizure patients [4,5] and to assess cognitive processes during neuroscience and psychology experiments [6,7]. In addition, EEG is used in brain computer interface devices, which enable command of an electronic device by brain activity modulation [8][9][10][11]. Current EEG systems can have as few as four electrodes [11] or as many as 256 electrodes.…”

Section: Introductionmentioning

confidence: 99%

How Many Electrodes Are Really Needed for EEG-Based Mobile Brain Imaging?

Lau¹,

Gwin²,

Ferris³

2012

JBBS

View full text Add to dashboard Cite

A noninvasive method for imaging the human brain during mobile activities could have far reaching benefits for studies of human motor control, for research and treatment of neurological disabilities, and for brain-controlled powered prosthetic limbs or orthoses. Several recent studies have demonstrated that electroencephalography (EEG) can be used to image the brain during locomotion provided that signal processing techniques, such as independent Component Analysis (ICA), are used to parse electrocortical activity from artifact contaminated EEG. However, these studies used high-density 256-channel EEG sensor arrays, which are likely too time-consuming to setup in a clinical or field setting. Therefore, it is important to evaluate how reducing the number of EEG channel signals affects the electrocortical source signals that can be parsed from EEG recorded during standing and walking while concurrently performing a visual oddball discrimination task. Specifically, we computed temporal and spatial correlations between electrocortical sources parsed from high-density EEG and electrocortical sources parsed from reduced-channel subsets of the original high-density EEG. For this task, our results indicate that on average an EEG montage with as few as 35 channels may be sufficient to record the two most dominate electrocortical sources (temporal and spatial R2 > 0.9). Correlations for additional electrocortical sources decreased linearly such that the least dominant sources extracted from the 35 channel dataset had temporal and spatial correlations of approximately 0.7. This suggests that for certain applications the number of EEG sensors used for mobile brain imaging could be vastly reduced, but researchers and clinicians must consider the expected distribution of relevant electrocortical sources when determining the number of EEG sensors necessary for a particular application.

show abstract

Machine learning for real-time single-trial EEG-analysis: From brain–computer interfacing to mental state monitoring

Cited by 419 publications

References 33 publications

Advanced Neurotechnologies for Chronic Neural Interfaces: New Horizons and Clinical Opportunities

Advanced Neurotechnologies for Chronic Neural Interfaces: New Horizons and Clinical Opportunities

Neurophysiological predictor of SMR-based BCI performance

How Many Electrodes Are Really Needed for EEG-Based Mobile Brain Imaging?

Contact Info

Product

Resources

About