Personal identification number application using brain-computer interface

We report the development of a new modality of personal identification application based on a brain-computer interface (BCI) framework. We used our previously developed visual stimulation in the proposed system, the overall architecture of which was designed to be simple for practical application. An electroencephalography amplifier with a single electrode channel (Oz) was sufficient for our system. A computationally inexpensive algorithm and small datasets were used in processing. Five healthy volunteers participated in offline performance evaluation experiments. Mean accuracy was 90% under the best conditions. Encouraged by these results, we plan to continue developing the BCI-based personal identification application for real-time systems.