Development of a cognitive BMI "neurocommunicator" as a communication aid of patients with severe motor deficits

Abstract

A cognitive brain-machine interface (BMI), "neurocommunicator" has been developed by the author's research group in AIST in order to support communication of patients with severer motor deficits. The system can identify candidate messages (pictograms) in real time from electroencephalography (EEG) data, combining three core technologies; 1) a portable/wireless EEG recorder; 2) a high-speed and high-accuracy decoding algorithm; and 3) a hierarchical message generation system. The accuracy of the model at single predictions of the target was generally over 95%, corresponding to about 32 bits per minute for normal subjects. Monitor experiments have been also started for patients at their home, in which further technical improvements are required.