Mapping forearm muscle activity with a single linear HD-EMG grid and muscle synergy model during hand-wrist movements

The characterization of the forearm muscle activation during hand-wrist movement is of key importance for allowing the development of control algorithms for prosthetic or assistive devices for the restoration of hand functions. Despite this importance, the state of the art solutions still fail to achieve stable performance due to the complex nature of the forearm muscles and the spatial superimpositions that are recorded when attempting EMG-based devices. The use of a high number of electrodes via HD-EMG systems, together with linear superimposition models coming from the muscle synergy theories, can represent a viable solution to this complexity, by identifying the coordination structures, instead of the single-channel activity. In this work, a simple hand-wrist movement protocol has been developed to test the possibility of identifying synergy-like structures and to link their activities to the performed movement. Results show that this solution, using optimized extraction algorithms, can yield a function-based description that can be directly linked to a set of basic hand-wrist gestures.Clinical relevance- This paper presents a novel methodology for characterizing gesture-dependent muscle activity in the forearm that can be employed for control and rehabilitation purposes.