Traditional Methods and Deep Learning in BCI-FES: A Review

Brain-computer interface (BCI) technology has gained significant attention in the field of neurological rehabilitation, particularly through the integration of BCI with functional electrical stimulation (FES), known as the BCI-FES system. However, existing reviews of BCI technology rarely delve into the specifics of algorithm development and its application within the BCI-FES domain. As such, the present paper aims to bridge this gap by providing an in-depth examination of the current advancements in electroencephalogram (EEG) signal processing algorithms for the FES area, highlighting recent developments and improvements in this area. Our review of the existing literature reveals that traditional algorithms continue to dominate the field of BCI-FES. Only a small number of deep learning algorithms have been successfully applied in practical BCI-FES systems. Because the ongoing development of these deep learning algorithms holds significant potential, they may represent a key direction for future research and system performance enhancement.