EEG-based classification of imaginary left and right foot movements using beta rebound

Abstract

ObjectiveThe purpose of this study was to investigate cortical lateralization of event-related (de)synchronization during left and right foot motor imagery tasks and to determine classification accuracy of the two imaginary movements in a brain–computer interface (BCI) paradigm. MethodsWe recorded 31-channel scalp electroencephalograms (EEGs) from nine healthy subjects during brisk imagery tasks of left and right foot movements. EEG was analyzed with time–frequency maps and topographies, and the accuracy rate of classification between left and right foot movements was calculated. ResultsBeta rebound at the end of imagination (increase of EEG beta rhythm amplitude) was identified from the two EEGs derived from the right-shift and left-shift bipolar pairs at the vertex. This process enabled discrimination between right or left foot imagery at a high accuracy rate (maximum 81.6% in single trial analysis). ConclusionThese data suggest that foot motor imagery has potential to elicit left–right differences in EEG, while BCI using the unilateral foot imagery can achieve high classification accuracy, similar to ordinary BCI, based on hand motor imagery. SignificanceBy combining conventional discrimination techniques, the left–right discrimination of unilateral foot motor imagery provides a novel BCI system that could control a foot neuroprosthesis or a robotic foot.