A multimodal electroencephalogram visualization system can promote stroke patients' motor imagery

Abstract

Objective To explore the effect of a multimodal electroencephalogram (EEG) data visualization system on the motor imagery ability of stroke survivors. Methods Twenty stroke patients were randomly divided into an experimental group and a control group, each of 10. Both groups were provided with brain-computer interface-based motor imagery (MI) training. At the same time, the experimental group was monitored and guided using an online, multimodal EEG data visualization system developed in our department. The classification accuracy (CA) and event-related desynchronization (ERD) of the 2 groups' motor imagery were compared before and after the treatment. Results Before the treatment, no significant differences in the average CA of MI were found between the experiment group (50.92±2.08) and the control group (49.35±4.20)(P>0.05). After the treatment, however, the experimental group's average CA had increased to (64.52±5.27), significantly higher than that of the control group (51.18±5.02). When the stroke patients imaged affected upper extremity movements, obvious ERD was observed in the α frequency around the bilateral central motor regions of both groups, especially in the experimental group, but without significant differences between the two groups. However, no significant changes were found in the ERD of the β waves of the two groups(P>0.05). Conclusion The proposed online multimodal electroencephalogram data visualization system can help stroke patients imagine movements actively. It is worth spreading in clinical practice. Key words: Electroencephalogram; Visualization; Stroke; Motor Imagery