Design of a BCI Controlled Serious Game for Concentration Training.

The goal of this work is to design a BCI that can then be used to control a serious game for concentration training. A 32-Electrode cap system was used, in addition to a bandpass and notch filter, t-SNE dimension reduction, standard deviation outlier detection as well as a SVM classifier. A maximum classification accuracy of 80% was achieved when using a four class classification system. Our BCI-Controlled Serious Game is viable and we thus plan to evaluate our application in a pilot test.