CLASSIFICATION OF ELECTROENCEPHALOGRAPHIC PATTERNS OF IMAGINED AND REAL MOVEMENTS BY ONE HAND FINGERS USING THE SUPPORT VECTORS METHOD
Abstract
Background. The article considers the possibility of distinguishing the electroencephalogram (EEG) patterns associated with real and imagined movements by the right hand fingers with the support vectors method (SVM) for use in the development of the «brain-computer» interface. Methods. Six healthy subjects performed the real and imaginary pressing by the thumb and forefinger of the right hand. The researchers analyzed EEG sensorimotor cortex (C3 and Cz) in the time window 1600 ms after 750 ms from the beginning of the test. Indications for the classification were generated according to the 1st trial signals and when summing up 3, 5, 10 and 20 trials of the same type. For classification were applied linear SVM and SVM based on radial basis function. Results. The average accuracy of movements classification exceeded statistically random threshold and increased with the number of trials (average 44.7±11.4 % when summing up 20 trials). Maximum classification accuracy using a linear SVM was 58.1±5.5 %, RBF SVM - 57.8±5.8 %. Recognition accuracy by 1 and 20 trials for SVM based on radial basis function was higher than that for the linear SVM. Conclusions. The authors show the possibility of distinguishing between EEG patterns of imagined movements by one hand fingers using SVM-classifier.
About the Authors
K. M. Sonkin
St. Petersburg State Polytechnical University
Russian Federation
Russian Federation
L. A. Stankevich
St. Petersburg State Polytechnical University
Russian Federation
Russian Federation
Yu. G. Khomenko
St. Petersburg State Polytechnical University; Institute of Human Brain named after N.P. Bekhtereva, RAS
Russian Federation
Russian Federation
Zh. V. Nagornova
Institute of Evolutionary Physiology and Biochemistry named after I.M. Sechenov, RAS; St. Petersburg Institute of Foreign Economic Relations of Economics and Law
Russian Federation
Russian Federation
N. V. Shemyakina
Institute of Evolutionary Physiology and Biochemistry named after I.M. Sechenov, RAS; St. Petersburg Institute of Foreign Economic Relations of Economics and Law
Russian Federation
Russian Federation
References
1. Каплан А.Я., Кочетова А.Г., Шишкин С.Л. [и др.]. Экспериментально-теоретические основания и практические реализации технологии «интерфейс мозг - компьютер» // Бюллетень сибирской медицины. 2013. Т. 12, № 2. С. 21-29.
2. Фролов А.А., Бирюкова Е.В., Бобров П.Д. [и др.]. Принципы нейрореабилитации, основанные на использовании интерфейса «мозг-компьютер» и биологически адекватного управления экзоскелетоном // Физиология человека. 2013. Т. 39, № 2. С. 99-113.
3. Сороко С.И. Нейрофизиологические механизмы индивидуальной адаптации человека в Антарктиде. Л.: Наука, 1984. 152 с.
4. Сороко С.И., Трубачев В.В. Нейрофизиологические и психофизиологические основы адаптивного биоуправления. СПб.: Политехника-сервис, 2010. 607 с.
5. Blankertz B., Tangermann M., Vidaurre C. [et al.]. Berlin brain-computer interface: non-medical uses of BCI technology // Frontiers in neuroscience. 2010. Vol. 4, No. 198. P. 17.
6. Chang C.-C., Lin C.-J. LIBSVM: a library for support vector machines // ACM Transactions on Intelligent Systems and Technology. 2011. URL: http://www.csie.ntu.edu.tw/~cjlin/libsvm (дата обращения 12.02.2014).
7. Kaiser V., Kreilinger A., Müller-Putz G.R., Neuper Ch. First steps toward a motor imagery based stroke BCI: new strategy to set up a classifier // Frontiers in Neuroscience. 2011. Vol. 5, No. 86. P. 10.
8. Lotte F., Congedo M., Lecuyer A. [et al.]. Review of classification algorithms for EEG-based brain-computer interfaces // Journal of Neural Engineering. 2007. Vol.4. P. 1-24.
9. Miller K.J., Zanos S., Fetz E.E. [et al.]. Decoupling the cortical power spectrum reveals real-time representation of individual finger movements in humans // The Journal of Neuroscience. 2009. Vol. 29 (10). P. 3132-3137.
10. Morash V., Bai O., Furlani S. [et al.]. Classifying EEG signals preceding right hand, left hand, tongue, and right foot movements and motor imageries // Clinical Neurophysiology 2008. Vol. 119. P. 2570.
11. Neuper C., Scherer R., Reiner M., Pfurtscheller G. Imagery of motor actions: differential effects of kinesthetic and visual-motor mode of imagery in singletrial EEG // Cognitive Brain Research. 2005. Vol. 25. P. 668-677.
12. Quandt F., Reichert C., Hinrichs H. [et al.]. Single trial discrimination of individual finger movements on one hand: A combined MEG and EEG study // Neuroimage. 2012. Vol. 59. P. 3316-3324.
13. Vapnik V.N. An overview of statistical learning theory // IEEE Transact. on Neural Networks. 1999. Vol. 10, No. 5. P. 988-999.
14. Wolpaw J.R., Birbaumer N., McFarland D.J. [et al.]. Brain-computer interfaces for communication and control // Clin. Neurophysiol. 2002. Vol. 113. P. 767-791.
15. Xiao R., Ding L. Evaluation of EEG features in decoding individual finger movements from one hand // Computational and Mathematical Methods in Medicine. 2013. Vol. 2013, Art. ID 243257. 10 p.
Review
For citations:
Sonkin K.M., Stankevich L.A., Khomenko Yu.G., Nagornova Zh.V., Shemyakina N.V. CLASSIFICATION OF ELECTROENCEPHALOGRAPHIC PATTERNS OF IMAGINED AND REAL MOVEMENTS BY ONE HAND FINGERS USING THE SUPPORT VECTORS METHOD. Pacific Medical Journal. 2014;(2):30-35. (In Russ.)
Views:
253
Email this article 