| Description |
Surface electromyography (sEMG) is a noninvasive way to decode motor intent for applications such as prosthesis control, virtual reality, and stroke or spinal cord rehabilitation. State-of-the-art methods of decoding motor intent used in myoelectric prostheses have limited capabilities, partially due to limited sEMG data. Furthermore, frequent changes in electrode alignment restrict the ability to decode motor intent reliably over an extended period of time. To address these challenges, we developed an easy-to-don, low-cost sleeve that can acquire robust and repeatable sEMG signals from 32 integrated electrodes. Coated brass snaps and grommets were embedded into a neoprene sleeve to serve as dry electrodes and as placement markers for consistent position over time, respectively. The sleeve is quick and easy to put on. For five subjects, donning the sleeve was significantly faster than for a previous prototype and orders of magnitude faster than current clinical approaches. The sleeve was also rated significantly more comfortable and easier to use. The new sleeve yielded improved control of a virtual prosthetic hand over the previous prototype, with significantly lower error in both intended and unintended movements and equivalent signal-to-noise ratio. The two sleeves were equivalent in consistency of electrode positioning over time. The high electrode density and consistent donning allow for high degree-of-freedom motor decodes that are stable over time. Future work includes optimizing electrode positioning to maximize useful information or reduce electrode count and modifying the design to integrate with a prosthetic socket. |
