Offline analysis of context contribution to ERP-based typing BCI performance

Umut Orhan, Deniz Erdogmus, Brian Roark, Barry Oken, Melanie Fried-Oken

Research output: Contribution to journalArticle

23 Citations (Scopus)

Abstract

Objective. We aim to increase the symbol rate of electroencephalography (EEG) based brain-computer interface (BCI) typing systems by utilizing context information. Approach. Event related potentials (ERP) corresponding to a stimulus in EEG can be used to detect the intended target of a person for BCI. This paradigm is widely utilized to build letter-by-letter BCI typing systems. Nevertheless currently available BCI typing systems still require improvement due to low typing speeds. This is mainly due to the reliance on multiple repetitions before making a decision to achieve higher typing accuracy. Another possible approach to increase the speed of typing while not significantly reducing the accuracy of typing is to use additional context information. In this paper, we study the effect of using a language model (LM) as additional evidence for intent detection. Bayesian fusion of an n-gram symbol model with EEG features is proposed, and a specifically regularized discriminant analysis ERP discriminant is used to obtain EEG-based features. The target detection accuracies are rigorously evaluated for varying LM orders, as well as the number of ERP-inducing repetitions. Main results. The results demonstrate that the LMs contribute significantly to letter classification accuracy. For instance, we find that a single-trial ERP detection supported by a 4-gram LM may achieve the same performance as using 3-trial ERP classification for the non-initial letters of words. Significance. Overall, the fusion of evidence from EEG and LMs yields a significant opportunity to increase the symbol rate of a BCI typing system.

Original languageEnglish (US)
Article number066003
JournalJournal of Neural Engineering
Volume10
Issue number6
DOIs
StatePublished - Dec 2013

Fingerprint

Brain-Computer Interfaces
Brain computer interface
Electroencephalography
Evoked Potentials
Language
Bioelectric potentials
Fusion reactions
Discriminant Analysis
Discriminant analysis
Target tracking
Decision Making

ASJC Scopus subject areas

  • Biomedical Engineering
  • Cellular and Molecular Neuroscience

Cite this

Offline analysis of context contribution to ERP-based typing BCI performance. / Orhan, Umut; Erdogmus, Deniz; Roark, Brian; Oken, Barry; Fried-Oken, Melanie.

In: Journal of Neural Engineering, Vol. 10, No. 6, 066003, 12.2013.

Research output: Contribution to journalArticle

@article{4f76d9b576ae4cf6927264574d6f0378,
title = "Offline analysis of context contribution to ERP-based typing BCI performance",
abstract = "Objective. We aim to increase the symbol rate of electroencephalography (EEG) based brain-computer interface (BCI) typing systems by utilizing context information. Approach. Event related potentials (ERP) corresponding to a stimulus in EEG can be used to detect the intended target of a person for BCI. This paradigm is widely utilized to build letter-by-letter BCI typing systems. Nevertheless currently available BCI typing systems still require improvement due to low typing speeds. This is mainly due to the reliance on multiple repetitions before making a decision to achieve higher typing accuracy. Another possible approach to increase the speed of typing while not significantly reducing the accuracy of typing is to use additional context information. In this paper, we study the effect of using a language model (LM) as additional evidence for intent detection. Bayesian fusion of an n-gram symbol model with EEG features is proposed, and a specifically regularized discriminant analysis ERP discriminant is used to obtain EEG-based features. The target detection accuracies are rigorously evaluated for varying LM orders, as well as the number of ERP-inducing repetitions. Main results. The results demonstrate that the LMs contribute significantly to letter classification accuracy. For instance, we find that a single-trial ERP detection supported by a 4-gram LM may achieve the same performance as using 3-trial ERP classification for the non-initial letters of words. Significance. Overall, the fusion of evidence from EEG and LMs yields a significant opportunity to increase the symbol rate of a BCI typing system.",
author = "Umut Orhan and Deniz Erdogmus and Brian Roark and Barry Oken and Melanie Fried-Oken",
year = "2013",
month = "12",
doi = "10.1088/1741-2560/10/6/066003",
language = "English (US)",
volume = "10",
journal = "Journal of Neural Engineering",
issn = "1741-2560",
publisher = "IOP Publishing Ltd.",
number = "6",

}

TY - JOUR

T1 - Offline analysis of context contribution to ERP-based typing BCI performance

AU - Orhan, Umut

AU - Erdogmus, Deniz

AU - Roark, Brian

AU - Oken, Barry

AU - Fried-Oken, Melanie

PY - 2013/12

Y1 - 2013/12

N2 - Objective. We aim to increase the symbol rate of electroencephalography (EEG) based brain-computer interface (BCI) typing systems by utilizing context information. Approach. Event related potentials (ERP) corresponding to a stimulus in EEG can be used to detect the intended target of a person for BCI. This paradigm is widely utilized to build letter-by-letter BCI typing systems. Nevertheless currently available BCI typing systems still require improvement due to low typing speeds. This is mainly due to the reliance on multiple repetitions before making a decision to achieve higher typing accuracy. Another possible approach to increase the speed of typing while not significantly reducing the accuracy of typing is to use additional context information. In this paper, we study the effect of using a language model (LM) as additional evidence for intent detection. Bayesian fusion of an n-gram symbol model with EEG features is proposed, and a specifically regularized discriminant analysis ERP discriminant is used to obtain EEG-based features. The target detection accuracies are rigorously evaluated for varying LM orders, as well as the number of ERP-inducing repetitions. Main results. The results demonstrate that the LMs contribute significantly to letter classification accuracy. For instance, we find that a single-trial ERP detection supported by a 4-gram LM may achieve the same performance as using 3-trial ERP classification for the non-initial letters of words. Significance. Overall, the fusion of evidence from EEG and LMs yields a significant opportunity to increase the symbol rate of a BCI typing system.

AB - Objective. We aim to increase the symbol rate of electroencephalography (EEG) based brain-computer interface (BCI) typing systems by utilizing context information. Approach. Event related potentials (ERP) corresponding to a stimulus in EEG can be used to detect the intended target of a person for BCI. This paradigm is widely utilized to build letter-by-letter BCI typing systems. Nevertheless currently available BCI typing systems still require improvement due to low typing speeds. This is mainly due to the reliance on multiple repetitions before making a decision to achieve higher typing accuracy. Another possible approach to increase the speed of typing while not significantly reducing the accuracy of typing is to use additional context information. In this paper, we study the effect of using a language model (LM) as additional evidence for intent detection. Bayesian fusion of an n-gram symbol model with EEG features is proposed, and a specifically regularized discriminant analysis ERP discriminant is used to obtain EEG-based features. The target detection accuracies are rigorously evaluated for varying LM orders, as well as the number of ERP-inducing repetitions. Main results. The results demonstrate that the LMs contribute significantly to letter classification accuracy. For instance, we find that a single-trial ERP detection supported by a 4-gram LM may achieve the same performance as using 3-trial ERP classification for the non-initial letters of words. Significance. Overall, the fusion of evidence from EEG and LMs yields a significant opportunity to increase the symbol rate of a BCI typing system.

UR - http://www.scopus.com/inward/record.url?scp=84888985678&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84888985678&partnerID=8YFLogxK

U2 - 10.1088/1741-2560/10/6/066003

DO - 10.1088/1741-2560/10/6/066003

M3 - Article

C2 - 24099944

AN - SCOPUS:84888985678

VL - 10

JO - Journal of Neural Engineering

JF - Journal of Neural Engineering

SN - 1741-2560

IS - 6

M1 - 066003

ER -