Verification of Adaptive Collection for Brain Computer Interface

To provide speech prostheses for individuals with severe communication impairments, brain computer interfaces (BCIs) using silent speech have been studied. I proposed adaptive collection, which divided brainwaves into smaller elements and verified them, for BCIs using silent. This paper verified the effect of adaptive collection in comparison to the conventional method. Brainwaves were obtained when four subjects imagined vocalization. In adaptive collection, shortening time length of brainwaves for common spatial patterns was effective because the state of brainwaves changes fast when a subject imagined vocalization. As a result, using the adaptive collection with 12 ms of the time length and 20 elements for classification, the classification accuracies were improved to 87–99% and the averaged classification accuracy was improved to 93% for the pairwise classification /a/ vs. /u/ in the case of 63 channels of EEG.