Biosignal control is right around the corner. This technology allows users to operate a device by decoding their intention from signals such as muscle activity or brain activity. Currently, Deep Learning is one of the most promising approaches to decoding the user’s intention from biosignals.
However, many issues need addressing to bring this method from the lab to the real world. In this article we systematically review the literature on the implementation and evaluation of biosignal control systems that use deep learning methods for decoding. From this analysis we identify gaps in the current knowledge on this topic and we highlight insights from existing literature. We then use these insights to formulate recommendations for designing, implementing and evaluation proof-of-concept control systems.
This review should enable researchers that are new to the field of biosignal control to get an overview of the scope and most important aspects of the technology. Researchers with experience in one of the related scientific fields should be able to use this as a starting point to solving the issues that are specific to their field.
Link to the open access article https://doi.org/10.1088/1741-2552/ac4f9a