Independent Component Analysis for Different Movements Detection in BCI Application Based on Sensorimotor Rhythms

This article discusses the use of independent component analysis (ICA) to detect different movements that are most significant for the brain-computer interface, in relation to the rehabilitation of post-stroke patients with upper limb paresis. The control signal for the brain-computer interface uses synchronization (desynchronization) of the sensorimotor rhythm (SMR) during motor imagery. ICA is used to increase the signal-to-noise ratio in the case of a mixed signal by separating activity sources from each other. A comparative analysis of the effectiveness of the evoked synchronization (desynchronization) reactions detection of the SMR in the left hemisphere in the original EEG signal in channel C3 and in the ICA extracted using 7 electrodes was performed. The results obtained by subjects who participated in the experiment for the first time and those who received short-term feedback training were also analyzed. It is shown that in experiments on motor imagery, the continuous rhythmic movement of the graspopening of the hand is more effective than a single continuous grasp.