Automatic frequency bands segmentation using statistical similarity for power spectrum density based brain computer interfaces

Tian Lan, Deniz Erdogmus, Misha Pavel, Santosh Mathan

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

3 Citations (Scopus)

Abstract

Power spectrum density (PSD) of electroencephalogram (EEG) signals is a widely used feature for Brain Computer Interfaces (BCI). Usually, PSD features are integrated over different frequency bands, such as delta, theta, alpha, beta, gamma, which are based on well-established interpretations of EEG signals in prior experimental and clinical contexts. However, these predefined frequency bands do not necessarily relate to the optimal features for various BCI applications. In this paper, we propose an alternative feature dimensionality reduction method, which automatically determines the optimal number and the range of frequency bands. We applied the proposed method on EEG classification in the context of Augmented Cognition (AugCog) using BCI. The experimental results show that the proposed method can extract more robust features than features manually extracted from predefined frequency bands.

Original languageEnglish (US)
Title of host publicationIEEE International Conference on Neural Networks - Conference Proceedings
Pages4650-4655
Number of pages6
StatePublished - 2006
EventInternational Joint Conference on Neural Networks 2006, IJCNN '06 - Vancouver, BC, Canada
Duration: Jul 16 2006Jul 21 2006

Other

OtherInternational Joint Conference on Neural Networks 2006, IJCNN '06
CountryCanada
CityVancouver, BC
Period7/16/067/21/06

Fingerprint

Brain computer interface
Power spectrum
Frequency bands
Electroencephalography

ASJC Scopus subject areas

  • Software

Cite this

Lan, T., Erdogmus, D., Pavel, M., & Mathan, S. (2006). Automatic frequency bands segmentation using statistical similarity for power spectrum density based brain computer interfaces. In IEEE International Conference on Neural Networks - Conference Proceedings (pp. 4650-4655). [1716745]

Automatic frequency bands segmentation using statistical similarity for power spectrum density based brain computer interfaces. / Lan, Tian; Erdogmus, Deniz; Pavel, Misha; Mathan, Santosh.

IEEE International Conference on Neural Networks - Conference Proceedings. 2006. p. 4650-4655 1716745.

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

Lan, T, Erdogmus, D, Pavel, M & Mathan, S 2006, Automatic frequency bands segmentation using statistical similarity for power spectrum density based brain computer interfaces. in IEEE International Conference on Neural Networks - Conference Proceedings., 1716745, pp. 4650-4655, International Joint Conference on Neural Networks 2006, IJCNN '06, Vancouver, BC, Canada, 7/16/06.
Lan T, Erdogmus D, Pavel M, Mathan S. Automatic frequency bands segmentation using statistical similarity for power spectrum density based brain computer interfaces. In IEEE International Conference on Neural Networks - Conference Proceedings. 2006. p. 4650-4655. 1716745
Lan, Tian ; Erdogmus, Deniz ; Pavel, Misha ; Mathan, Santosh. / Automatic frequency bands segmentation using statistical similarity for power spectrum density based brain computer interfaces. IEEE International Conference on Neural Networks - Conference Proceedings. 2006. pp. 4650-4655
@inproceedings{e75a891347304a52bdb3ee7b24a227be,
title = "Automatic frequency bands segmentation using statistical similarity for power spectrum density based brain computer interfaces",
abstract = "Power spectrum density (PSD) of electroencephalogram (EEG) signals is a widely used feature for Brain Computer Interfaces (BCI). Usually, PSD features are integrated over different frequency bands, such as delta, theta, alpha, beta, gamma, which are based on well-established interpretations of EEG signals in prior experimental and clinical contexts. However, these predefined frequency bands do not necessarily relate to the optimal features for various BCI applications. In this paper, we propose an alternative feature dimensionality reduction method, which automatically determines the optimal number and the range of frequency bands. We applied the proposed method on EEG classification in the context of Augmented Cognition (AugCog) using BCI. The experimental results show that the proposed method can extract more robust features than features manually extracted from predefined frequency bands.",
author = "Tian Lan and Deniz Erdogmus and Misha Pavel and Santosh Mathan",
year = "2006",
language = "English (US)",
isbn = "0780394909",
pages = "4650--4655",
booktitle = "IEEE International Conference on Neural Networks - Conference Proceedings",

}

TY - GEN

T1 - Automatic frequency bands segmentation using statistical similarity for power spectrum density based brain computer interfaces

AU - Lan, Tian

AU - Erdogmus, Deniz

AU - Pavel, Misha

AU - Mathan, Santosh

PY - 2006

Y1 - 2006

N2 - Power spectrum density (PSD) of electroencephalogram (EEG) signals is a widely used feature for Brain Computer Interfaces (BCI). Usually, PSD features are integrated over different frequency bands, such as delta, theta, alpha, beta, gamma, which are based on well-established interpretations of EEG signals in prior experimental and clinical contexts. However, these predefined frequency bands do not necessarily relate to the optimal features for various BCI applications. In this paper, we propose an alternative feature dimensionality reduction method, which automatically determines the optimal number and the range of frequency bands. We applied the proposed method on EEG classification in the context of Augmented Cognition (AugCog) using BCI. The experimental results show that the proposed method can extract more robust features than features manually extracted from predefined frequency bands.

AB - Power spectrum density (PSD) of electroencephalogram (EEG) signals is a widely used feature for Brain Computer Interfaces (BCI). Usually, PSD features are integrated over different frequency bands, such as delta, theta, alpha, beta, gamma, which are based on well-established interpretations of EEG signals in prior experimental and clinical contexts. However, these predefined frequency bands do not necessarily relate to the optimal features for various BCI applications. In this paper, we propose an alternative feature dimensionality reduction method, which automatically determines the optimal number and the range of frequency bands. We applied the proposed method on EEG classification in the context of Augmented Cognition (AugCog) using BCI. The experimental results show that the proposed method can extract more robust features than features manually extracted from predefined frequency bands.

UR - http://www.scopus.com/inward/record.url?scp=40649114245&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=40649114245&partnerID=8YFLogxK

M3 - Conference contribution

SN - 0780394909

SN - 9780780394902

SP - 4650

EP - 4655

BT - IEEE International Conference on Neural Networks - Conference Proceedings

ER -