The article A. Gardetto, G. R. Müller-Putz, K. R. Eberlin, F. Bassetto, D. J. Atkins, M. Turri, G. Peternell, O. Neuper, and J. Ernst, “Restoration of Genuine Sensation and Proprioception of Individual Fingers Following Transradial Amputation with Targeted Sensory Reinnervation as a Mechanoneural Interface,” Journal of Clinical Medicine, vol. 14, iss. 2, p. 417, 2025 [1] [PDF, distributed as CC BY 4.0] describes a method similar to targeted muscle reinnervation (TMR) they termed targeted sensory reinnervation (TSR) and sensory function in arm stumps after specific surgery.
From what is known via all sorts of anecdotal channels and own experience, practical solutions are all the rage for prosthetic arms in real life, and, TMR/TRS appear to have their own set of issues.
One issue may be persistent pain after surgery. Another issue seems to be a maximally precise placement of the electrodes.
For TMR/TSR to work, muscle packs or nerves are surgically placed under the skin, next to each other. Electrodes then are placed on the skin, exactly over these muscle packs.
The trick is to get the electrodes to stay there, exactly there, because otherwise the electrodes cannot read the signals from the nerves or muscles there. If that does not work to perfection across the whole range of real life use of such a system, then the surgery itself may not turn out as useful as it could be hoped to be. As these rerouted nerve or muscle locations may be tightly spaced and, in themselves, small size wise, the metric aspect of the whole thing is of essence.
With that, it is surprising that the arm stump photographs with the pencil drawn innervation areas on the skin [1] do not show a scale bar in what we assume to be a scientific paper. What is interesting is a map that shows how large or small the useful area for signal elicitation is, on the skin. Such a map alongside a metric scale would be relevant to indicate how much slack there is for the socket to encase the stump. From the images, I estimate these electrode placements to allow for a placement error not in excess of 2 mm.
Only a bit of socket or electrode shift is needed to stop the method from reliably working. Thus reducing any shift is of essence. That, actually, is even already an issue with two electrode sites in conventional myoelectric hard sockets; there, the stump is immobilized maximally so the electrodes stay in a relative fixed position.
The result is a significantly reduced elbow range of motion and uncomfortable strain pain in the arm and shoulder, as I experienced with my hard socket for conventional myoelectric control [link]. Similar stories were seemingly traded on user forums where TMR caused real issues when the socket shifted or slipped even just a little. Given that this is practically relevant, relative socket / stump shifts need to be measured, understood, taken into account and the remedy must be comfortable.
With all that, sweat remains a significant issue, and with that, control error and other myoelectric issues [link, link, link].
[Bibtex]
@article{gardetto2025restoration,
title={Restoration of Genuine Sensation and Proprioception of Individual Fingers Following Transradial Amputation with Targeted Sensory Reinnervation as a Mechanoneural Interface},
author={Gardetto, Alexander and M{\"u}ller-Putz, Gernot R and Eberlin, Kyle R and Bassetto, Franco and Atkins, Diane J and Turri, Mara and Peternell, Gerfried and Neuper, Ortrun and Ernst, Jennifer},
journal={Journal of Clinical Medicine},
volume={14},
number={2},
pages={417},
year={2025},
publisher={MDPI}
}