The design of spatial selection using CUR decomposition to improve common spatial pattern for multi-trial EEG classification

Hilman Fauzi, Mohd Ibrahim Shapiai, Rubiyah Yusof, Gerard Bastiaan Remijn, Noor Akhmad Setiawan, Zuwairie Ibrahim

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

3 Citations (Scopus)

Abstract

The most important factor in EEG signal processing is the determination of relevant features in encoding the meaning of the signal. Obtaining relevant features for EEG can be done using a spatial filter. The Common Spatial Pattern (CSP) is known to produce discriminative features when processing EEG signals. Yet, CSP is also sensitive to noise and is channel-dependent, as it is considered to be a spatial filter. However, the disadvantage of CSP is that channels containing only noise are also considered as active channels. In this paper, the design of a filter for spatial selection is proposed using CUR decomposition to select important channels or the time segment of EEG trials in order to improve CSP performance. CUR decomposition can also be used as a noise rejection technique because CUR can be used in factorizing the given EEG signals. In other words, CUR decomposition rejects the non-active channels, which typically contain noise, before spatially filtering the EEG signals. Once the EEG signal is decomposed based on the importance of the channels, time segmentation, and EEG factorization, the decomposed signal can be used as input to the CSP. In general, three approaches were proposed in this framework: (1) channel selection, i.e., C selection; (2) time segment selection, R; and (3) signal factorization, U. Furthermore, the performance accuracy between the original CSP and CSP in which the input was spatially filtered by the proposed framework was validated using datasets IVa of BCI competition III. The test results show that the CSP with spatial selection using C selection and U factorization offers 12% and 9% improvement compared to the original CSP, respectively. Hence, the proposed method in this study can be used as a spatial filter to improve the CSP performance.

Original languageEnglish
Title of host publicationModeling, Design and Simulation of Systems - 17th Asia Simulation Conference, AsiaSim 2017, Proceedings
EditorsMohamed Sultan Mohamed Ali, Herman Wahid, Nurul Adilla Mohd Subha, Shafishuhaza Sahlan, Mohd Amri Md. Yunus, Ahmad Ridhwan Wahap
PublisherSpringer Verlag
Pages428-442
Number of pages15
ISBN (Print)9789811064623
DOIs
Publication statusPublished - Jan 1 2017
Event17th International Conference on Asia Simulation, AsiaSim 2017 - Melaka, Malaysia
Duration: Aug 27 2017Aug 29 2017

Publication series

NameCommunications in Computer and Information Science
Volume751
ISSN (Print)1865-0929

Other

Other17th International Conference on Asia Simulation, AsiaSim 2017
CountryMalaysia
CityMelaka
Period8/27/178/29/17

Fingerprint

Spatial Pattern
Electroencephalography
Decomposition
Decompose
Factorization
Filter
Design
Electroencephalogram
Signal processing
Rejection
Signal Processing
Encoding
Segmentation
Filtering
Processing
Dependent

All Science Journal Classification (ASJC) codes

  • Computer Science(all)
  • Mathematics(all)

Cite this

Fauzi, H., Shapiai, M. I., Yusof, R., Remijn, G. B., Setiawan, N. A., & Ibrahim, Z. (2017). The design of spatial selection using CUR decomposition to improve common spatial pattern for multi-trial EEG classification. In M. S. Mohamed Ali, H. Wahid, N. A. Mohd Subha, S. Sahlan, M. A. Md. Yunus, & A. R. Wahap (Eds.), Modeling, Design and Simulation of Systems - 17th Asia Simulation Conference, AsiaSim 2017, Proceedings (pp. 428-442). (Communications in Computer and Information Science; Vol. 751). Springer Verlag. https://doi.org/10.1007/978-981-10-6463-0_37

The design of spatial selection using CUR decomposition to improve common spatial pattern for multi-trial EEG classification. / Fauzi, Hilman; Shapiai, Mohd Ibrahim; Yusof, Rubiyah; Remijn, Gerard Bastiaan; Setiawan, Noor Akhmad; Ibrahim, Zuwairie.

Modeling, Design and Simulation of Systems - 17th Asia Simulation Conference, AsiaSim 2017, Proceedings. ed. / Mohamed Sultan Mohamed Ali; Herman Wahid; Nurul Adilla Mohd Subha; Shafishuhaza Sahlan; Mohd Amri Md. Yunus; Ahmad Ridhwan Wahap. Springer Verlag, 2017. p. 428-442 (Communications in Computer and Information Science; Vol. 751).

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

Fauzi, H, Shapiai, MI, Yusof, R, Remijn, GB, Setiawan, NA & Ibrahim, Z 2017, The design of spatial selection using CUR decomposition to improve common spatial pattern for multi-trial EEG classification. in MS Mohamed Ali, H Wahid, NA Mohd Subha, S Sahlan, MA Md. Yunus & AR Wahap (eds), Modeling, Design and Simulation of Systems - 17th Asia Simulation Conference, AsiaSim 2017, Proceedings. Communications in Computer and Information Science, vol. 751, Springer Verlag, pp. 428-442, 17th International Conference on Asia Simulation, AsiaSim 2017, Melaka, Malaysia, 8/27/17. https://doi.org/10.1007/978-981-10-6463-0_37
Fauzi H, Shapiai MI, Yusof R, Remijn GB, Setiawan NA, Ibrahim Z. The design of spatial selection using CUR decomposition to improve common spatial pattern for multi-trial EEG classification. In Mohamed Ali MS, Wahid H, Mohd Subha NA, Sahlan S, Md. Yunus MA, Wahap AR, editors, Modeling, Design and Simulation of Systems - 17th Asia Simulation Conference, AsiaSim 2017, Proceedings. Springer Verlag. 2017. p. 428-442. (Communications in Computer and Information Science). https://doi.org/10.1007/978-981-10-6463-0_37
Fauzi, Hilman ; Shapiai, Mohd Ibrahim ; Yusof, Rubiyah ; Remijn, Gerard Bastiaan ; Setiawan, Noor Akhmad ; Ibrahim, Zuwairie. / The design of spatial selection using CUR decomposition to improve common spatial pattern for multi-trial EEG classification. Modeling, Design and Simulation of Systems - 17th Asia Simulation Conference, AsiaSim 2017, Proceedings. editor / Mohamed Sultan Mohamed Ali ; Herman Wahid ; Nurul Adilla Mohd Subha ; Shafishuhaza Sahlan ; Mohd Amri Md. Yunus ; Ahmad Ridhwan Wahap. Springer Verlag, 2017. pp. 428-442 (Communications in Computer and Information Science).
@inproceedings{6e7e4fa1422b4dfaa185baecbbb5fbd2,
title = "The design of spatial selection using CUR decomposition to improve common spatial pattern for multi-trial EEG classification",
abstract = "The most important factor in EEG signal processing is the determination of relevant features in encoding the meaning of the signal. Obtaining relevant features for EEG can be done using a spatial filter. The Common Spatial Pattern (CSP) is known to produce discriminative features when processing EEG signals. Yet, CSP is also sensitive to noise and is channel-dependent, as it is considered to be a spatial filter. However, the disadvantage of CSP is that channels containing only noise are also considered as active channels. In this paper, the design of a filter for spatial selection is proposed using CUR decomposition to select important channels or the time segment of EEG trials in order to improve CSP performance. CUR decomposition can also be used as a noise rejection technique because CUR can be used in factorizing the given EEG signals. In other words, CUR decomposition rejects the non-active channels, which typically contain noise, before spatially filtering the EEG signals. Once the EEG signal is decomposed based on the importance of the channels, time segmentation, and EEG factorization, the decomposed signal can be used as input to the CSP. In general, three approaches were proposed in this framework: (1) channel selection, i.e., C selection; (2) time segment selection, R; and (3) signal factorization, U. Furthermore, the performance accuracy between the original CSP and CSP in which the input was spatially filtered by the proposed framework was validated using datasets IVa of BCI competition III. The test results show that the CSP with spatial selection using C selection and U factorization offers 12{\%} and 9{\%} improvement compared to the original CSP, respectively. Hence, the proposed method in this study can be used as a spatial filter to improve the CSP performance.",
author = "Hilman Fauzi and Shapiai, {Mohd Ibrahim} and Rubiyah Yusof and Remijn, {Gerard Bastiaan} and Setiawan, {Noor Akhmad} and Zuwairie Ibrahim",
year = "2017",
month = "1",
day = "1",
doi = "10.1007/978-981-10-6463-0_37",
language = "English",
isbn = "9789811064623",
series = "Communications in Computer and Information Science",
publisher = "Springer Verlag",
pages = "428--442",
editor = "{Mohamed Ali}, {Mohamed Sultan} and Herman Wahid and {Mohd Subha}, {Nurul Adilla} and Shafishuhaza Sahlan and {Md. Yunus}, {Mohd Amri} and Wahap, {Ahmad Ridhwan}",
booktitle = "Modeling, Design and Simulation of Systems - 17th Asia Simulation Conference, AsiaSim 2017, Proceedings",
address = "Germany",

}

TY - GEN

T1 - The design of spatial selection using CUR decomposition to improve common spatial pattern for multi-trial EEG classification

AU - Fauzi, Hilman

AU - Shapiai, Mohd Ibrahim

AU - Yusof, Rubiyah

AU - Remijn, Gerard Bastiaan

AU - Setiawan, Noor Akhmad

AU - Ibrahim, Zuwairie

PY - 2017/1/1

Y1 - 2017/1/1

N2 - The most important factor in EEG signal processing is the determination of relevant features in encoding the meaning of the signal. Obtaining relevant features for EEG can be done using a spatial filter. The Common Spatial Pattern (CSP) is known to produce discriminative features when processing EEG signals. Yet, CSP is also sensitive to noise and is channel-dependent, as it is considered to be a spatial filter. However, the disadvantage of CSP is that channels containing only noise are also considered as active channels. In this paper, the design of a filter for spatial selection is proposed using CUR decomposition to select important channels or the time segment of EEG trials in order to improve CSP performance. CUR decomposition can also be used as a noise rejection technique because CUR can be used in factorizing the given EEG signals. In other words, CUR decomposition rejects the non-active channels, which typically contain noise, before spatially filtering the EEG signals. Once the EEG signal is decomposed based on the importance of the channels, time segmentation, and EEG factorization, the decomposed signal can be used as input to the CSP. In general, three approaches were proposed in this framework: (1) channel selection, i.e., C selection; (2) time segment selection, R; and (3) signal factorization, U. Furthermore, the performance accuracy between the original CSP and CSP in which the input was spatially filtered by the proposed framework was validated using datasets IVa of BCI competition III. The test results show that the CSP with spatial selection using C selection and U factorization offers 12% and 9% improvement compared to the original CSP, respectively. Hence, the proposed method in this study can be used as a spatial filter to improve the CSP performance.

AB - The most important factor in EEG signal processing is the determination of relevant features in encoding the meaning of the signal. Obtaining relevant features for EEG can be done using a spatial filter. The Common Spatial Pattern (CSP) is known to produce discriminative features when processing EEG signals. Yet, CSP is also sensitive to noise and is channel-dependent, as it is considered to be a spatial filter. However, the disadvantage of CSP is that channels containing only noise are also considered as active channels. In this paper, the design of a filter for spatial selection is proposed using CUR decomposition to select important channels or the time segment of EEG trials in order to improve CSP performance. CUR decomposition can also be used as a noise rejection technique because CUR can be used in factorizing the given EEG signals. In other words, CUR decomposition rejects the non-active channels, which typically contain noise, before spatially filtering the EEG signals. Once the EEG signal is decomposed based on the importance of the channels, time segmentation, and EEG factorization, the decomposed signal can be used as input to the CSP. In general, three approaches were proposed in this framework: (1) channel selection, i.e., C selection; (2) time segment selection, R; and (3) signal factorization, U. Furthermore, the performance accuracy between the original CSP and CSP in which the input was spatially filtered by the proposed framework was validated using datasets IVa of BCI competition III. The test results show that the CSP with spatial selection using C selection and U factorization offers 12% and 9% improvement compared to the original CSP, respectively. Hence, the proposed method in this study can be used as a spatial filter to improve the CSP performance.

UR - http://www.scopus.com/inward/record.url?scp=85029001380&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85029001380&partnerID=8YFLogxK

U2 - 10.1007/978-981-10-6463-0_37

DO - 10.1007/978-981-10-6463-0_37

M3 - Conference contribution

AN - SCOPUS:85029001380

SN - 9789811064623

T3 - Communications in Computer and Information Science

SP - 428

EP - 442

BT - Modeling, Design and Simulation of Systems - 17th Asia Simulation Conference, AsiaSim 2017, Proceedings

A2 - Mohamed Ali, Mohamed Sultan

A2 - Wahid, Herman

A2 - Mohd Subha, Nurul Adilla

A2 - Sahlan, Shafishuhaza

A2 - Md. Yunus, Mohd Amri

A2 - Wahap, Ahmad Ridhwan

PB - Springer Verlag

ER -