A microcontroller-based ambulatory brain-computer interface training system

Brain-computer interface (BCI) systems are generally developed as desktop devices that confine a subject to the training site and limit training protocol options. We hypothesize that a BCI system that can be used at any time and place will improve current training methods and will thus accelerate the realization of a human's potential towards controlling an external device. For this reason, we have developed an ambulatory BCI (ABCI) training system. It consists of a microcontroller-based circuit that acquires, digitizes, and processes up to two amplified EEG signals and then transmits the bandpass-limited power to a PDA for classification, translation, and training game control. The game allows the user to train by providing a visual feedback signal that is to be cognitively controlled in a task-oriented challenge. Evaluation of the ABCI shows that the accuracy and response time are equivalent to a desktop BCI system. The ABCI can therefore be used chronically and away from any training site, thus removing barriers inherent in traditional BCI systems. This should allow more advanced training protocols to be developed and implemented in order to more effectively study the ability of the brain to adapt and control augmentative devices.