Use of graph metrics to classify motor imagery based BCI

The major aim of this work was to propose a novel method to perform a motor imagery (MI) based brain control interfacing system (BCI) classification using a single feature derived from the graph theory applied to connectivity measures. In particular, the characterization of small world coefficient is studied along different scenarios. Two connectivity measures as phase locking value (PLV) and coherence, two different frequency bands and two different time slots division (static and 3 different time windows). The second objective of this work was to study the viability of a novel stimuli for using on MI based BCIs, emotional schematic faces. Two emotions were showed to the participants: happiness and sadness to perform their MI tasks. Accuracy rates of up to 91.1% suggest that this is a promising strategy for BCI classifiers.