Abstract
Noninvasive brain–computer interfaces (BCI) translate subject’s electroencephalogram (EEG) features into device commands. Large feature sets should be down-selected for efficient feature translation. This work proposes two different feature down-selection algorithms for BCI: (a) a sequential forward selection; and (b) an across-group variance. Power rar ratios (PRs) were extracted from the EEG data for movement imagery discrimination. Event-related potentials (ERPs) were employed in the discrimination of cue-evoked responses. While center-out arrows, commonly used in calibration sessions, cued the subjects in the first experiment (for both PR and ERP analyses), less stimulating arrows that were centered in the visual field were employed in the second experiment (for ERP analysis). The proposed algorithms outperformed other three popular feature selection algorithms in movement imagery discrimination. In the first experiment, both algorithms achieved classification errors as low as 12.5% reducing the feature set dimensionality by more than 90%. The classification accuracy of ERPs dropped in the second experiment since centered cues reduced the amplitude of cue-evoked ERPs. The two proposed algorithms effectively reduced feature dimensionality while increasing movement imagery discrimination and detected cue-evoked ERPs that reflect subject attention.
Similar content being viewed by others
References
Babiloni C et al (1999) Human movement-related potentials vs desynchronization of EEG alpha rhythm: a high-resolution EEG study. NeuroImage 10:658–665
Bashashati A, Ward RK, Birch GE (2005) A new design of the asynchronous brain–computer interface using the knowledge of the path of features. In: Proc 2nd IEEE-EMBS conference on neural engineering, Arlington, VA, pp 101–104
Boostani R et al (2007) A comparison approach toward finding the best feature and classifier in cue-based BCI. Med Biol Eng Comput 45:403–412
Courville T, Thompson B (2001) Use of structure coefficients in published multiple regression articles: β is not enough. Educ Psychol Meas 61:229–248
Dias NS et al (2009) Feature Down-Selection in brain–computer Interfaces. In: Proc. of the 4th international IEEE EMBS conference on neural engineering. Antalya, Turkey, pp 323–326
Dillon WR, Mulani N, Frederick DG (1989) On the use of component scores in the presence of group structure. J Cons Res 16:106–112
Duda RO, Hart PE, Stork DG (2001) Pattern classification. Wiley, New York
Fabiani GE et al (2004) Conversion of EEG activity into cursor movement by a brain–computer interface (BCI). IEEE Trans Neural Syst Rehabil Eng 12:331–338
Fatourechi M et al (2006) Automatic user customization for improving the performance of a self-paced brain interface. Med Biol Eng Comput 44:1093–1104
Flury B (1997) A first course in multivariate statistics. Springer, New York
Grafton ST et al (1997) Premotor cortex activation during observation and naming of familiar tools. Neuroimage 6:231–236
Guger C et al (2001) Rapid prototyping of an EEG-based brain–computer interface (BCI). IEEE Trans Neural Syst Rehabil Eng 9:49–58
Guo F et al (2008) A brain–computer interface using motion-onset visual evoked potential. J Neural Eng 5:477–485
Guyon I, Eliseeff A (2003) An introduction to variable and feature selection. J Mach Learn Res 3:1157–1182
Hillyard SA, Vogel EK, Luck SJ (1998) Sensory gain control (amplification) as a mechanism of selective attention: electrophysiological and neuroimaging evidence. Phil Trans R Soc Lond B 353:1257–1270
Jolliffe IT (2002) Principal component analysis. Springer, New York
Klemm M, Haueisen J, Ivanova G (2009) Independent component analysis: comparison of algorithms for the investigation of surface electrical brain activity. Med Biol Eng Comput 47:413–423
Kruglikov SY, Schiff SJ (2003) Interplay of EEG phase and auditory evoked neural activity. J Neurosci 2:10122–10127
Krusienski DJ et al (2008) Toward enhanced P300 speller performance. J Neurosci Meth 167:15–21
Lal TN et al (2004) Support vector channel selection in BCI. IEEE Trans Biomed Eng 51:1003–1010
Lee P-L et al (2008) Brain computer interface using flash onset and offset visual evoked potentials. Clin Neurophysiol 119:605–616
Liao X et al (2007) Combining spatial filters for the classification of singal-trial EEG in a finger movement task. IEEE Trans Biomed Eng 54:821–831
Luck SJ (2005) An introduction to the event-related potential technique. The MIT Press, Cambridge, MA
Millán J et al (2002) Relevant EEG features for the classification of spontaneous motor-related tasks. Biol Cybern 86:89–95
Müller T et al (2000) Selecting relevant electrode positions for classification tasks based on the electro-encephalogram. Med Biol Eng Comput 38:62–67
Pfurtscheller G, Neuper C (2001) Motor imagery and direct brain–computer communication. Proc IEEE 89:1123–1134
Schiff SJ (2005) Dangerous phase. Neuroinformatics 3:315–318
Schiff SJ et al (2005) Neuronal spatiotemporal pattern discrimination: the dynamical evolution of seizures. Neuroimage 28:1043–1055
Sun S, Zhang C (2006) Adaptive feature extraction for EEG signal classification. Med Biol Eng Comput 44:931–935
Wang Y, Makeig S (2009) Predicting intended movement direction using EEG from human posterior parietal cortex. In: Schmorrow DD et al (eds) Augmented cognition, HCII 2009. LNAI 5638, pp 437–446
Wascher E, Wauschkuhn B (1996) The interaction of stimulus- and response-related processes measured by event-related lateralizations of the EEG. Electroencephalogr Clin Neurophysiol 99:149–162
Wolpaw JR et al (2002) Brain–computer interfaces for communication and control. Clin Neurophysiol 113:767–791
Wolpaw JR, McFarland DJ (2004) Control of a two-dimensional movement signal by a noninvasive brain–computer interface in humans. Proc Natl Acad Sci USA 101:17849–17854
Yu L, Liu H (2004) Efficient feature selection via analysis of relevance and redundancy. J Mach Learn 5:1205–1224
Acknowledgments
The project described was supported by Award Number K25NS061001 from the US National Institute Of Neurological Disorders And Stroke. N. S. Dias was supported by the Portuguese Foundation for Science and Technology under the grant SFRH/BD/21529/2005. S. J. Schiff was supported by the NIH grant K02MH01493, The Pennsylvania Keystone Innovation Zone Program and The Pennsylvania Tobacco Settlement. The authors acknowledge the contribution of L.R. Jacinto on method implementation.
Author information
Authors and Affiliations
Corresponding author
Appendix
Appendix
This appendix describes the calculation of the discrimination coefficients a employed in:
Considering that z is the discrimination function and Y is the feature matrix. Initially, the SVD of the within-group covariance matrix W is calculated as:
S is a diagonal matrix, and U appears twice since covariance matrices are symmetric. B is the between-group covariance matrix. In order to obtain a better coordinate system, the vector a in the following Fisher criterion:
is replaced by \( {\mathbf{US}}^{{ - {1/2}}} {\mathbf{U}}^{\rm T} {\mathbf{v}}, \) resulting in:
In general, for a symmetric matrix H, the maximum of \( {\mathbf{v}}^{\rm T} {\mathbf{Hv}} \) is attained for the first singular vector v = v 1. Similarly, the maximization of α may be calculated through the following SVD:
The maximum of α is \( {\mathbf{v}}_{1}^{T} {\mathbf{VHV}}^{T} {\mathbf{v}}_{1} = {\varvec{\lambda}}_{1} , \) is the highest singular value of H. Converting back to original coordinates a:
which is equivalent to the first column of \( {\mathbf{US}}^{{ - {1/2}}} {\mathbf{U}}^{\rm T} {\mathbf{V}}. \)
Rights and permissions
About this article
Cite this article
Dias, N.S., Kamrunnahar, M., Mendes, P.M. et al. Feature selection on movement imagery discrimination and attention detection. Med Biol Eng Comput 48, 331–341 (2010). https://doi.org/10.1007/s11517-010-0578-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11517-010-0578-1