A pairwise output coding method for multi-class EEG classification of a self-induced BCI

Abstract

In brain computer interface (BCI) research, electroencephalography (EEG) is themost widely used method due to its noninvasiveness, high temporal resolution andportability. Most of the EEG-based BCI studies are aimed at developingmethodologies for signal processing, feature extraction and classification. In thisstudy, an experimental EEG study was carried out with six subjects performingimagery mental and motor tasks. We present a multi-class EEG decoding with anovel pairwise output coding method of EEGs to improve the performance of selfinducedBCI systems. This method involves an augmented one-versus-onemulticlass classification with less time and reduced number of electrodes.Furthermore, a train repetition number is introduced in the training step to optimizethe data selection. The difference among right and left hemispheres is alsosearched. Finally, the difference between experienced and novice subjects is alsoobserved.The experimental results have demonstrated that, the use of proposed classificationalgorithm produces high classification accuracies (98%) with nine channels.Reduced numbers of channels (four channels) have 100% accuracies for mentaltasks and 87% accuracies for motor tasks with Support Vector Machines (SVM).The classification accuracies are quite high though the proposed one-versus-onetechnique worked well compared to the classical method. The results would bepromising for a real-time study.