Semi-supervised feature extraction for EEG classification

Abstract

Two semi-supervised feature extraction methods are proposed for electroencephalogram (EEG) classification. They aim to alleviate two important limitations in brain–computer interfaces (BCIs). One is on the requirement of small training sets owing to the need of short calibration sessions. The second is the time-varying property of signals, e.g., EEG signals recorded in the training and test sessions often exhibit different discriminant features. These limitations are common in current practical applications of BCI systems and often degrade the performance of traditional feature extraction algorithms. In this paper, we propose two strategies to obtain semi-supervised feature extractors by improving a previous feature extraction method extreme energy ratio (EER). The two methods are termed semi-supervised temporally smooth EER and semi-supervised importance weighted EER, respectively. The former constructs a regularization term on the preservation of the temporal manifold of test samples and adds this as a constraint to the learning of spatial filters. The latter defines two kinds of weights by exploiting the distribution information of test samples and assigns the weights to training data points and trials to improve the estimation of covariance matrices. Both of these two methods regularize the spatial filters to make them more robust and adaptive to the test sessions. Experimental results on data sets from nine subjects with comparisons to the previous EER demonstrate their better capability for classification.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3

References

  1. 1.

    Vaughan TM, Heetderks WJ, Trejo LJ, Rymer WZ, Weinrich M, Moore MM, Kübler A, Dobkin BH, Birbaumer N, Donchin E, Wolpaw EW, Wolpaw JR (2003) Brain-computer interface technology: a review of the second international meeting. IEEE Trans Neural Syst Rehabil Eng 11(2): 94–109

    Article  Google Scholar 

  2. 2.

    Wolpaw JR, Birbaumer N, McFarland DJ, Pfurtscheller G, Vaughan TM (2002) Brain–computer interfaces for communication and control. Clinical Neurophysiol 113(6): 767–791

    Article  Google Scholar 

  3. 3.

    Jain AK, Duin RPW, Mao J (2000) Statistical pattern recognition: a review. IEEE Trans Pattern Anal Mach Intell 22(1): 4–37

    Article  Google Scholar 

  4. 4.

    Sun S (2010) Extreme energy difference for feature extraction of EEG signals. Expert Syst Appl 37(6): 4350–4357

    Article  Google Scholar 

  5. 5.

    Krauledat M, Tangermann M, Blankertz B, Müller KR (2008) Towards zero training for brain-computer interfacing. PLoS One 3(8): e2967

    Article  Google Scholar 

  6. 6.

    Shenoy, P, Krauledat M, Blankertz B, Rao RPN, Müller KR (2006) Towards adaptive classification for BCI. J Neural Eng 3: R13–R23

    Article  Google Scholar 

  7. 7.

    Millán JR (2004) On the need for on-Line learning in brain–computer interfaces. IEEE Int Conf Neural Netw—Conference Proceedings 4: 2877–2882

    Google Scholar 

  8. 8.

    Vidaurre C, and Schlogl A, Cabeza R, Scherer R, Pfurtscheller G (2007) Study of on-line adaptive discriminant analysis for EEG-based brain computer interfaces. IEEE Trans Biomed Eng 54(3): 550–556

    Article  Google Scholar 

  9. 9.

    Sun S, Zhang C (2006) Adaptive feature extraction for EEG signal classification. Med Biol Eng Comput 44(10): 931–935

    Article  Google Scholar 

  10. 10.

    Li Y, Guan C, Li H, Chin Z (2008) A self-training semi-supervised SVM algorithm and its application in an EEG-based brain computer interface speller system. Pattern Recogn Lett 29(9):1285–1294

    Article  Google Scholar 

  11. 11.

    Li Y, Guan C (2008) Joint feature re-extraction and classification using an iterative semi-supervised support vector machine algorithm. Mach Learn 71(1): 33–53

    MathSciNet  Article  Google Scholar 

  12. 12.

    Zhu X (2006) Semi-supervised learning literature survey. Computer Science, University of Wisconsin-Madison

  13. 13.

    Song Y, Zhang C, Lee J, Wang F, Xiang S, Zhang D (2009) Semi-supervised discriminative classification with application to tumorous tissues segmentation of MR brain images. Pattern Anal Appl 12(2): 99–115

    MathSciNet  Article  Google Scholar 

  14. 14.

    Xie B, Mu Y, Tao D, Huang K (2011) m-SNE: Multiview stochastic neighbor embedding. Trans Syst Man Cybern Part B 41(4): 1088–1096

    Article  Google Scholar 

  15. 15.

    Xia T, Tao D, Mei T, Zhang Y (2010) Multiview spectral embedding. IEEE Trans Syst Man Cybern Part B 40(6): 1438–1446

    Article  Google Scholar 

  16. 16.

    Belkin M, and Niyogi P (2004) Semi-supervised learning on Riemannian manifolds. Mach Learn 56(1): 209–239

    MATH  Article  Google Scholar 

  17. 17.

    Lee H, Yoo J, Choi S (2010) Semi-supervised nonnegative matrix factorization. IEEE Signal Process Lett 17(1): 4–7

    Article  Google Scholar 

  18. 18.

    Sun S (2008) The extreme energy ratio criterion for EEG feature extraction. Lect Notes Comput Sci 5164: 919–928

    Article  Google Scholar 

  19. 19.

    Guan N, Tao D, Luo Z, Yuan B (2011) Manifold regularized discriminative nonnegative matrix factorization with fast gradient descent. IEEE Trans Image Process 20(7): 2030–2048

    MathSciNet  Article  Google Scholar 

  20. 20.

    Wang X, Tao D, Li Z (2011) Subspaces indexing model on Grassmann manifold for image search. IEEE Trans Image Process 20(9): 2627–2635

    MathSciNet  Article  Google Scholar 

  21. 21.

    Wang X, Tao D, Li Z (2010) Entropy controlled Laplacian regularization for least square regression. Signal Process 90(6): 2043–2049

    MATH  Article  Google Scholar 

  22. 22.

    Hill NJ, Lal TN, Schröder M, Hinterberger T, Widman G, Elger CE, Schölkopf B, Birbaumer N (2006) Classifying event-related desynchronization in EEG, ECoG and MEG Signal. Lect Notes Comput Sci 4174:404–413

    Article  Google Scholar 

  23. 23.

    Kittler JV, Young PC (1973) A new approach to feature selection based on the Karhunen-Loève expansion. Pattern Recogn 5: 335–352

    MathSciNet  Article  Google Scholar 

  24. 24.

    Kim, D. and Sra, S. and Dhillon, I.S. (2007) Fast Newton-type methods for the least squares nonnegative matrix approximation problem. Proceedings of IEEE International Conference on Data Mining, 343–354

  25. 25.

    Kim H, Park H (2008) Non-negative matrix factorization based on alternating non-negativity constrained least squares and active set method. SIAM J Matrix Anal Appl 30(2): 713–730

    MathSciNet  MATH  Article  Google Scholar 

  26. 26.

    Guan N, Tao D, Luo Z, Yuan B (2012) Online nonnegative matrix factorization with robust stochastic approximation. IEEE Trans Neural Netw Learn Syst 23(7): 1087–1099

    Article  Google Scholar 

  27. 27.

    Guan N, Tao D, Luo Z, Yuan B (2012) An optimal gradient method for nonnegative matrix factorization. IEEE Trans Signal Process 60(6): 2882–2898

    MathSciNet  Article  Google Scholar 

  28. 28.

    Kim, J. and Park, H. (2012) Toward faster nonnegative matrix factorization: a new algorithm and comparisons. Proceedings of IEEE International Conference on Data Mining, 353–362

  29. 29.

    Millán JR (2008) Robust common spatial patterns for EEG signal preprocessing. Proceedings of the 30th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2087–2090

  30. 30.

    Belkin M, Niyogi P, Sindhwani V (2006) Manifold regularization: a geometric framework for learning from labeled and unlabeled examples. J Mach Learn Res 7: 2399–2434

    MathSciNet  MATH  Google Scholar 

  31. 31.

    He X, Niyogi P (2003) Locality preserving projections. In: advances in neural information processing systems 16, MIT Press, Cambridge, MA

  32. 32.

    Müller KR, Anderson CW, Birch GE (2003) Linear and non-linear methods for brain-computer interfaces. IEEE Trans Neural Syst Rehabil Eng 11(2): 165–169

    Article  Google Scholar 

  33. 33.

    Sugiyama M, Kanamori T, Suzuki T, Hido S, Sese J, Takeuchi I, Wang L (2009) A density-ratio framework for statistical data processing. Inform Media Technol 4(4): 962–987

    Google Scholar 

  34. 34.

    Sugiyama M, Suzuki T, Nakajima S, Kashima H, von Bünau P, Kawanabe M (2008) Direct importance estimation for covariate shift adaptation. Annals Inst Stat Math 60(4): 699–746

    MATH  Article  Google Scholar 

  35. 35.

    Sajda P, Gerson A, Mller KR, Blankertz B, Parra L (2003) A data analysis competition to evaluatemachine learning algorithms for use in brai computer interfaces. IEEE Trans Neural Syst Rehabil Eng 11(2): 184–185

    Article  Google Scholar 

  36. 36.

    Nunez PL, Srinivasan R, Westdorp AF, Wijesinghe DM, Tucker RB, Cadusch PJ (1997) EEG coherency I: Statistics, reference electrode, volume conduction, laplacians, cortical imaging, and interpretation at multiple scale. Electroenceph. Clinical Neurophysiol 103: 499–515

    Article  Google Scholar 

  37. 37.

    Müller-Gerking J, Pfurtscheller G, Flyvbjerg H (1999) Designing optimal spatial filters for single-trial EEG classification in a movement task. Clinical Neurophysiol 110(5): 787–798

    Article  Google Scholar 

  38. 38.

    Cai D, He X, Han J (2007) Semi-supervised discriminant analysis. Proceedings of the IEEE 11th International Conference on Computer Vision 110(5): 787–798

  39. 39.

    Parra LC, Spence CD, Gerson AD, Sajda P (2005) Recipes for the linear analysis of EEG. NeuroImage 28: 326–341

    Article  Google Scholar 

  40. 40.

    Müller KR, Mika S, Ratsch G, Tsuda K, Schölkopf B (2001) An introduction to kernel based learning algorithms. IEEE Trans Neural Netw 12: 181–201

    Article  Google Scholar 

  41. 41.

    Tian X, Tao D, Rui Y (2011) Sparse transfer learning for interactive video search reranking. CoRR abs/1103.2756. 2011.

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Wenting Tu.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Tu, W., Sun, S. Semi-supervised feature extraction for EEG classification. Pattern Anal Applic 16, 213–222 (2013). https://doi.org/10.1007/s10044-012-0298-2

Download citation

Keywords

  • Semi-supervised learning
  • Feature extraction EEG classification
  • Extreme energy ratio
  • Regularization
  • Density ratio