Decoding hand and wrist movement intention from chronic stroke survivors with hemiparesis using a user-friendly, wearable EMG-based neural interface.
Eric C Meyers, David Gabrieli, Nick Tacca and 5 others
PMID 38218901WHAT IT FOUND
A wearable EMG sleeve decoded 12 hand and wrist movements in seven chronic stroke survivors with 77.1% accuracy.
Decoding worked best when visible movement was present. For severe impairments, binary intent detection achieved 86.7% accuracy, though no physical therapy outcome was tested.
Key findings
01The neural network decoder achieved 77.1% accuracy across all 12 movements in stroke survivors, dropping to 27.3% when no visible movement occurred.
02For individuals with severe hand impairment, a binary classifier distinguishing rest from movement achieved 86.7% accuracy, suggesting a single-DOF control pathway.
03Two participants successfully used the system for real-time closed-loop control of a virtual hand, with decoding inference times under 100 ms.
STILL TO COME
How it was doneWhat they foundWhat it means for PTsWhat it means for OTs
Read the rest of this summary
You get three full summaries a month, free, and we do not ask for a card. Search, the TL;DRs and your library stay unlimited either way.
What it does not show
The study was a pilot with only seven stroke survivors, limiting generalisability. Decoding was performed offline or with a virtual hand, not with a physical assistive device like an exoskeleton or FES system, so the impact of device interference on EMG signals was not tested. Participants kept their elbow stationary on a table and did not interact with objects, which may not reflect real-world functional use. The study did not measure clinical outcomes such as motor recovery, dexterity, or quality of life; it only measured decoding accuracy. Able-bodied controls were not age-matched to the stroke group, which may have influenced the comparison of EMG signal characteristics.
Declared interests
The study was supported by Battelle Memorial Institute, where the NeuroLife system is developed. Able-bodied subjects were employees of Battelle. The authors declare this as research support from a non-U.S. government source, but the affiliation with the device developer is a potential conflict of interest.
The easy way to misread this
Do not assume this system is ready for clinical use or that it improves motor recovery. The study validated decoding accuracy in a laboratory setting with a virtual hand, not functional independence with a real assistive device. The high accuracy in severe impairment (86.7%) applies only to a binary rest/move task, not to complex grasp control.