Removal of ocular artifacts from EEG using learned templates

Max Quinn, Santosh Mathan, Misha Pavel

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)

Abstract

Electroencephalogram (EEG) data can provide information on cognitive states and processes with high temporal resolution, but to take full advantage of this temporal resolution, common transients such as blinks and eye movements must be accounted for without censoring data. This can require additional hardware, large amounts of data, or manual inspection. In this paper we introduce a greedy, template-based method for modeling and removing transient activity. The method iteratively models an input and updates a template; a process which quickly converges to a unique and efficient approximation of the input. When combined with standard source separation techniques such as Independent Component Analysis (ICA) or Principal Component Analysis (PCA), the method shows promise for the automatic and data driven removal of ocular artifacts from EEG data. In this paper we outline our method, provide evidence for its effectiveness using synthetic EEG data, and demonstrate its effect on real EEG data recorded as part of a minimally constrained cognitive task.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Pages371-380
Number of pages10
Volume8027 LNAI
DOIs
StatePublished - 2013
Externally publishedYes
Event7th International Conference on Foundations of Augmented Cognition, AC 2013, Held as Part of 15th International Conference on Human-Computer Interaction, HCI International 2013 - Las Vegas, NV, United States
Duration: Jul 21 2013Jul 26 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8027 LNAI
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other7th International Conference on Foundations of Augmented Cognition, AC 2013, Held as Part of 15th International Conference on Human-Computer Interaction, HCI International 2013
CountryUnited States
CityLas Vegas, NV
Period7/21/137/26/13

Fingerprint

Electroencephalography
Template
Source separation
Eye movements
Independent component analysis
Principal component analysis
Source Separation
Eye Movements
Inspection
Independent Component Analysis
Censoring
Hardware
Data-driven
Principal Component Analysis
Electroencephalogram
Update
Converge
Approximation
Modeling
Demonstrate

Keywords

  • BCI
  • EEG
  • EOG
  • ICA
  • matching pursuit
  • PCA

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Quinn, M., Mathan, S., & Pavel, M. (2013). Removal of ocular artifacts from EEG using learned templates. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8027 LNAI, pp. 371-380). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8027 LNAI). https://doi.org/10.1007/978-3-642-39454-6_39

Removal of ocular artifacts from EEG using learned templates. / Quinn, Max; Mathan, Santosh; Pavel, Misha.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8027 LNAI 2013. p. 371-380 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8027 LNAI).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Quinn, M, Mathan, S & Pavel, M 2013, Removal of ocular artifacts from EEG using learned templates. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 8027 LNAI, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8027 LNAI, pp. 371-380, 7th International Conference on Foundations of Augmented Cognition, AC 2013, Held as Part of 15th International Conference on Human-Computer Interaction, HCI International 2013, Las Vegas, NV, United States, 7/21/13. https://doi.org/10.1007/978-3-642-39454-6_39
Quinn M, Mathan S, Pavel M. Removal of ocular artifacts from EEG using learned templates. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8027 LNAI. 2013. p. 371-380. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-642-39454-6_39
Quinn, Max ; Mathan, Santosh ; Pavel, Misha. / Removal of ocular artifacts from EEG using learned templates. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8027 LNAI 2013. pp. 371-380 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{18e38ab3c19346fbb0996943dfd466d7,
title = "Removal of ocular artifacts from EEG using learned templates",
abstract = "Electroencephalogram (EEG) data can provide information on cognitive states and processes with high temporal resolution, but to take full advantage of this temporal resolution, common transients such as blinks and eye movements must be accounted for without censoring data. This can require additional hardware, large amounts of data, or manual inspection. In this paper we introduce a greedy, template-based method for modeling and removing transient activity. The method iteratively models an input and updates a template; a process which quickly converges to a unique and efficient approximation of the input. When combined with standard source separation techniques such as Independent Component Analysis (ICA) or Principal Component Analysis (PCA), the method shows promise for the automatic and data driven removal of ocular artifacts from EEG data. In this paper we outline our method, provide evidence for its effectiveness using synthetic EEG data, and demonstrate its effect on real EEG data recorded as part of a minimally constrained cognitive task.",
keywords = "BCI, EEG, EOG, ICA, matching pursuit, PCA",
author = "Max Quinn and Santosh Mathan and Misha Pavel",
year = "2013",
doi = "10.1007/978-3-642-39454-6_39",
language = "English (US)",
isbn = "9783642394539",
volume = "8027 LNAI",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "371--380",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Removal of ocular artifacts from EEG using learned templates

AU - Quinn, Max

AU - Mathan, Santosh

AU - Pavel, Misha

PY - 2013

Y1 - 2013

N2 - Electroencephalogram (EEG) data can provide information on cognitive states and processes with high temporal resolution, but to take full advantage of this temporal resolution, common transients such as blinks and eye movements must be accounted for without censoring data. This can require additional hardware, large amounts of data, or manual inspection. In this paper we introduce a greedy, template-based method for modeling and removing transient activity. The method iteratively models an input and updates a template; a process which quickly converges to a unique and efficient approximation of the input. When combined with standard source separation techniques such as Independent Component Analysis (ICA) or Principal Component Analysis (PCA), the method shows promise for the automatic and data driven removal of ocular artifacts from EEG data. In this paper we outline our method, provide evidence for its effectiveness using synthetic EEG data, and demonstrate its effect on real EEG data recorded as part of a minimally constrained cognitive task.

AB - Electroencephalogram (EEG) data can provide information on cognitive states and processes with high temporal resolution, but to take full advantage of this temporal resolution, common transients such as blinks and eye movements must be accounted for without censoring data. This can require additional hardware, large amounts of data, or manual inspection. In this paper we introduce a greedy, template-based method for modeling and removing transient activity. The method iteratively models an input and updates a template; a process which quickly converges to a unique and efficient approximation of the input. When combined with standard source separation techniques such as Independent Component Analysis (ICA) or Principal Component Analysis (PCA), the method shows promise for the automatic and data driven removal of ocular artifacts from EEG data. In this paper we outline our method, provide evidence for its effectiveness using synthetic EEG data, and demonstrate its effect on real EEG data recorded as part of a minimally constrained cognitive task.

KW - BCI

KW - EEG

KW - EOG

KW - ICA

KW - matching pursuit

KW - PCA

UR - http://www.scopus.com/inward/record.url?scp=84884879429&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84884879429&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-39454-6_39

DO - 10.1007/978-3-642-39454-6_39

M3 - Conference contribution

SN - 9783642394539

VL - 8027 LNAI

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 371

EP - 380

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

ER -