In this paper, we develop a new type of brain-computer interface (BCI) which is able to control a computer game by motor imagery electroencephalogram (EEG). We propose a new framework of feature extractions using common spatial frequency patterns (CSFP) for classification of motor imagery EEG. The aim of our BCI system is to provide an on-line "hit rat" game control with short response time and subject-specific adaptation of system parameters. Our BCI system is able to detect three different motor imagery-related brain patterns (imagination of limb movements: left hand, right hand and both feet) from the ongoing brain activity by using only five EEG channels. The best hit accuracy of the game with fast response time attained by subject 2 is about 73%, which demonstrates that our BCI system has the ability of providing much fast BCI of even 1 s per command.
CITATION STYLE
Zhao, Q., Zhang, L., & Li, J. (2008). Multi-Task BCI for Online Game Control. In Autonomous Systems – Self-Organization, Management, and Control (pp. 29–37). Springer Netherlands. https://doi.org/10.1007/978-1-4020-8889-6_4
Mendeley helps you to discover research relevant for your work.