A Feature Filter for EEG Using Cycle-GAN Structure

Yue Yao; Jo Plested; Tom Gedeon

doi:10.1007/978-3-030-04239-4_51

A Feature Filter for EEG Using Cycle-GAN Structure

Yue Yao^*, Jo Plested, Tom Gedeon

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

6 Citations (Scopus)

Abstract

The brain-computer interface (BCI) has become one of the most important biomedical research fields and has created many useful applications. As an important component of BCI, electroencephalography (EEG) is in general sensitive to noise and rich in all kinds of information from our brain. In this paper, we introduce a new strategy to filter out unwanted features from EEG signals using GAN-based autoencoders. Filtering out signals relating to one property of the EEG signal while retaining another is similar to the way we can listen to just one voice during a party. This approach has many potential applications including in privacy and security. We use the UCI EEG dataset on alcoholism for our experiments. Our experiment results show that our novel GAN based structure can filter out alcoholism information for 66% of EEG signals with an average of only 6.2% accuracy lost.

Original language	English
Title of host publication	Neural Information Processing - 25th International Conference, ICONIP 2018, Proceedings
Editors	Seiichi Ozawa, Andrew Chi Sing Leung, Long Cheng
Publisher	Springer Verlag
Pages	567-576
Number of pages	10
ISBN (Print)	9783030042387
DOIs	https://doi.org/10.1007/978-3-030-04239-4_51
Publication status	Published - 2018
Event	25th International Conference on Neural Information Processing, ICONIP 2018 - Siem Reap, Cambodia Duration: 13 Dec 2018 → 16 Dec 2018

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	11307 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	25th International Conference on Neural Information Processing, ICONIP 2018
Country/Territory	Cambodia
City	Siem Reap
Period	13/12/18 → 16/12/18

Access to Document

10.1007/978-3-030-04239-4_51

Cite this

Yao, Y., Plested, J., & Gedeon, T. (2018). A Feature Filter for EEG Using Cycle-GAN Structure. In S. Ozawa, A. C. S. Leung, & L. Cheng (Eds.), Neural Information Processing - 25th International Conference, ICONIP 2018, Proceedings (pp. 567-576). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11307 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-030-04239-4_51

Yao, Yue ; Plested, Jo ; Gedeon, Tom. / A Feature Filter for EEG Using Cycle-GAN Structure. Neural Information Processing - 25th International Conference, ICONIP 2018, Proceedings. editor / Seiichi Ozawa ; Andrew Chi Sing Leung ; Long Cheng. Springer Verlag, 2018. pp. 567-576 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{241040c742f945959a593885147359c1,

title = "A Feature Filter for EEG Using Cycle-GAN Structure",

abstract = "The brain-computer interface (BCI) has become one of the most important biomedical research fields and has created many useful applications. As an important component of BCI, electroencephalography (EEG) is in general sensitive to noise and rich in all kinds of information from our brain. In this paper, we introduce a new strategy to filter out unwanted features from EEG signals using GAN-based autoencoders. Filtering out signals relating to one property of the EEG signal while retaining another is similar to the way we can listen to just one voice during a party. This approach has many potential applications including in privacy and security. We use the UCI EEG dataset on alcoholism for our experiments. Our experiment results show that our novel GAN based structure can filter out alcoholism information for 66% of EEG signals with an average of only 6.2% accuracy lost.",

keywords = "Brain-Computer interface, Deep learning, EEG, Generative adversarial nets, Image translation",

author = "Yue Yao and Jo Plested and Tom Gedeon",

note = "Publisher Copyright: {\textcopyright} 2018, Springer Nature Switzerland AG.; 25th International Conference on Neural Information Processing, ICONIP 2018 ; Conference date: 13-12-2018 Through 16-12-2018",

year = "2018",

doi = "10.1007/978-3-030-04239-4_51",

language = "English",

isbn = "9783030042387",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "567--576",

editor = "Seiichi Ozawa and Leung, {Andrew Chi Sing} and Long Cheng",

booktitle = "Neural Information Processing - 25th International Conference, ICONIP 2018, Proceedings",

address = "Germany",

}

Yao, Y, Plested, J & Gedeon, T 2018, A Feature Filter for EEG Using Cycle-GAN Structure. in S Ozawa, ACS Leung & L Cheng (eds), Neural Information Processing - 25th International Conference, ICONIP 2018, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 11307 LNCS, Springer Verlag, pp. 567-576, 25th International Conference on Neural Information Processing, ICONIP 2018, Siem Reap, Cambodia, 13/12/18. https://doi.org/10.1007/978-3-030-04239-4_51

A Feature Filter for EEG Using Cycle-GAN Structure. / Yao, Yue; Plested, Jo; Gedeon, Tom.
Neural Information Processing - 25th International Conference, ICONIP 2018, Proceedings. ed. / Seiichi Ozawa; Andrew Chi Sing Leung; Long Cheng. Springer Verlag, 2018. p. 567-576 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 11307 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - A Feature Filter for EEG Using Cycle-GAN Structure

AU - Yao, Yue

AU - Plested, Jo

AU - Gedeon, Tom

PY - 2018

Y1 - 2018

N2 - The brain-computer interface (BCI) has become one of the most important biomedical research fields and has created many useful applications. As an important component of BCI, electroencephalography (EEG) is in general sensitive to noise and rich in all kinds of information from our brain. In this paper, we introduce a new strategy to filter out unwanted features from EEG signals using GAN-based autoencoders. Filtering out signals relating to one property of the EEG signal while retaining another is similar to the way we can listen to just one voice during a party. This approach has many potential applications including in privacy and security. We use the UCI EEG dataset on alcoholism for our experiments. Our experiment results show that our novel GAN based structure can filter out alcoholism information for 66% of EEG signals with an average of only 6.2% accuracy lost.

AB - The brain-computer interface (BCI) has become one of the most important biomedical research fields and has created many useful applications. As an important component of BCI, electroencephalography (EEG) is in general sensitive to noise and rich in all kinds of information from our brain. In this paper, we introduce a new strategy to filter out unwanted features from EEG signals using GAN-based autoencoders. Filtering out signals relating to one property of the EEG signal while retaining another is similar to the way we can listen to just one voice during a party. This approach has many potential applications including in privacy and security. We use the UCI EEG dataset on alcoholism for our experiments. Our experiment results show that our novel GAN based structure can filter out alcoholism information for 66% of EEG signals with an average of only 6.2% accuracy lost.

KW - Brain-Computer interface

KW - Deep learning

KW - EEG

KW - Generative adversarial nets

KW - Image translation

UR - http://www.scopus.com/inward/record.url?scp=85059030710&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-04239-4_51

DO - 10.1007/978-3-030-04239-4_51

M3 - Conference contribution

SN - 9783030042387

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

SP - 567

EP - 576

BT - Neural Information Processing - 25th International Conference, ICONIP 2018, Proceedings

A2 - Ozawa, Seiichi

A2 - Leung, Andrew Chi Sing

A2 - Cheng, Long

PB - Springer Verlag

T2 - 25th International Conference on Neural Information Processing, ICONIP 2018

Y2 - 13 December 2018 through 16 December 2018

ER -

Yao Y, Plested J, Gedeon T. A Feature Filter for EEG Using Cycle-GAN Structure. In Ozawa S, Leung ACS, Cheng L, editors, Neural Information Processing - 25th International Conference, ICONIP 2018, Proceedings. Springer Verlag. 2018. p. 567-576. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-04239-4_51

A Feature Filter for EEG Using Cycle-GAN Structure

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this