It is shown that despite the fact that the motor imagery based brain computer interface does not rely on any particular feature of EEG signal defined a priori, system designed on the basis of EEG signal classifier is indeed controlled by the signals originating in the motor cortex. To prove this the most distinguishable EEG patterns were extracted by means of Independent Component Analysis with consequent cross-validation procedure used to select the independent components significant to the brain computer interface performance. Sources of the brain activity represented by the chosen independent components were located using single dipole approximation with individual head geometry model. These sources were found in the bottom of the central sulcus, area 3a, for each subject. These results are in good agreement with the outcome of fMRI study conducted under the same conditions. © 2012 Springer-Verlag.
CITATION STYLE
Frolov, A. A., Húsek, D., Bobrov, P. D., Korshakov, A., Chernikova, L., Konovalov, R., & Mokienko, O. (2012). Localizing sources of brain activity relevant to motor imagery brain-computer interface performance, using individual head geometry. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7367 LNCS, pp. 369–378). https://doi.org/10.1007/978-3-642-31346-2_42
Mendeley helps you to discover research relevant for your work.