(18) ICAとダイポール解析によるsingle-trial EEG-based BCIの基礎的検討
第21回生体・生理工学シンポジウム論文集
(平成18年11月17日〜19日 鹿児島かごしま県民交流センター),pp263-264,2006-11
In order to develop a new brain-computer interface (BCI) using EEG, independent
component analysis (ICA) and then dipole analysis were applied to scalp-recorded
single-trial EEG during movement imagery (MI). Up to the 20th independent
component (IC), each of them contained one or two positive (or negative)
peaks, and these peaks were analyzed by 1- or 2-dipole source localization
method. Some of the obtained dipoles were localized to contralateral motor
and premotor cortices in addition to ipsilateral ones. The ratios of the
right to left dipoles was significantly different between right- and left-handed
MI tasks for one right-handed male subject, contrary to the results of Qin
et al. (2004).