Classification of EEG Signals Using Filter Bank Common Spatial Pattern Based on Fisher and Laplacian Criteria

Abstract:

Common spatial pattern (CSP) is a highly successful algorithm in motor imagery based brain-computer interfaces (BCIs). The performance of the algorithm, however, depends largely on the operational frequency bands. To address the problem, a filter bank was applied to find optimal frequency bands. In filter bank, CSP was applied in all sub-band signals for feature extraction. The feature selection is the key of filter bank method for increasing classification performance. In this study, coefficient decimation (CD) technique was used to devise filter bank, while Fisher score and Laplacian score were proposed as feature selection criterion. In off-line analysis, the proposed method yielded relatively better cross-validation classification accuracies.

• Periodical: Applied Mechanics and Materials (Volumes 239-240)
• Pages: 1033-1038
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.239-240.1033
• Online since: December 2012
• Authors: Qing Guo Wei, Bin Wan, Zong Wu Lu
• Keywords: Brain Computer Interface (BCI), Coefficient Decimation, Feature Selection, Filter Bank

References

[1] J.R. Wolpaw, N. Birbaumer, D.J. McFarland, et al., Clini. Neurophysio., 2002, 113: 767-791.

[2] G. Pfurtscheeler, A. Brunner, A. Schlogl, et al., Neurolmages, 2006, 31 (1): 153-159.

[3] B. Blankertz, R. Tomioka, S. Lemm, et al., IEEE Signal Processing Magazine, 2008, 25: 41-56.

[4] G. Pfurtscheller, A. Stancak, G. Edlinger, Electroence. Clin. Neurophysio., 1997, 102: 316-325.

[5] S. Lemm, B. Blankertz, G. Curio, et al., IEEE Trans. Biomed. Eng., 2005, 52 (9): 1541-1548.

[6] G. Dornhege, B. Blankertz, M. Krauledat, et al., IEEE Trans. Biomed. Eng., 2006, 53: 274-281.

[7] Q. Novi, C. Guan, T.H. Huy, et al, Sub-band common spatial pattern (SBCSP) for brain-computer interface, in Proc. of IEEE EMBS, Hawaii, USA, May 2-5, 2007, pp.219-225.

DOI: 10.1109/cne.2007.369647

[8] K.K. Ang, Z.Y. Chin, H. Zhang, et al, Filter bank common spatial pattern (FBCSP) in brain-computer interface, in Proc. of IEEE Int. Joint Conf. on Neural Networks, Hong Kong, China, June 2008, pp.2390-2397.

DOI: 10.1109/ijcnn.2008.4634130

[9] R. Mahesh, A.P. Vinod, Coefficient decimation approach for realizing reconfigurable finite impulse response filters, in Proc. of IEEE ISCAS, Seattle, USA, May 2008, pp.81-84.

DOI: 10.1109/iscas.2008.4541359

[10] J. Muller-Gerking, G. Pfurtscheller, H. Flyvbjerg, Clin. Neurophysio., 1999, 110: 787-798.

[11] H. Ramoser, J. Muller-Gerking, G. Pfurtscheller, IEEE Trans. Rehabil. Eng., 2000, 8: 441-446.

[12] R. Kohavi and G. John, Artificail Intelligence, 1997, 97: 273-324.

[13] R. Duda, P. Hart, D Stork, Pattern Classification, 2nd edn, Wiley, New York, (2001).

[14] X. He, D. Cai and P. Niyogi, Advance in Neural Info. Proc. Sys., vol. 18, MIT Press, (2005).