Channel-Selection-Based Right and Left-Hand Movement Detection: A Step Toward Online EEG Processing

The interpretation of brain signals has consistently been a captivating subject for rehabilitation specialists. This interest prompts the development of interfaces that facilitate the interface between human cognition and an external device. These interfaces are commonly regarded as brain-computer interfaces (BCIs). However, a significant limitation of many existing BCI systems is their inadequate computational efficiency, which hinders their practical applicability in real-world scenarios. In this study, we present a channel-selection-based electroencephalogram (EEG) analyzing method, which enhances the BCIs to be used in real-time applications. To be specific, this method enables real-time systems to utilize Independent Component Analysis (ICA) by reducing channel numbers to three channels and selecting the most suitable set of channels to apply ICA. According to the results, the implementation of this approach leads to a marginal reduction (around 5%) in ultimate accuracy. However, this trade-off is accompanied by a substantial decrease in computational time. Consequently, such a modification holds the potential to enhance BCI functionality and facilitate the design of more intricate neurofeedback systems for Motor Imagery (MI) tasks.