Steady-State Visual Evoked Potential Classifiers for their Use in Industrial Brain-Computer Interfaces.

Compared to brain-computer interfaces (BCIs) developed in the laboratory settings, real-world BCIs are generally constrained by several factors: low resolution of the signal, its low signal-to-noise ratio and a lack of well-labeled data. In these conditions, good performance of BCI classifiers become even more challenging. In the current project, we have set up an EEG study to enrich the repertoire of BCI classifiers of steady-state visual evoked potentials (SSVEPs) and adapt them to industrial and clinical settings. In the early state of the project, we have tested existing solutions, such as deep convolutional neural network, canonical correlation analysis (CCA) and task-related component analysis (TRCA). Based on the performance of each algorithm, analysis of their perspectives was made, and the program of further research was outlined to create adaptive and flexible repertoire of SSVEP classifiers.