Neural Noise- Implanted microelectrode arrays in reinnervated muscles allow separation of neural drives from transferred polyfunctional nerves.1
- Intramuscular recording enables decomposition of mixed neural drives for prosthetic control.1
- The approach informs reinnervation strategies and neuroprosthetic control.1 1
Weekly enrichment (2026-07-20)
- Three men with upper-limb amputation (ages 62, 34 and 52; two transhumeral, one glenohumeral) who had undergone targeted muscle reinnervation (TMR) 7–12 years earlier took part, with four reinnervated muscles studied—three reinnervated by the ulnar nerve and one by the radial nerve.23
- Each muscle was recorded with a single implanted intramuscular microelectrode array of 40 sites (140 µm electrode diameter) linearly distributed at 500 µm spacing over a 2-cm span.3
- Blind-source separation decomposed the multichannel intramuscular EMG into individual motor-unit spike trains, yielding 111 motor units across the four muscles at a pulse-to-noise ratio above 30 dB.3
- Average identified motor units per task were 7.6 ± 0.8, 4.6 ± 1.1, 3.1 ± 1.4 and 3.8 ± 1.5 for the four muscles, and baseline noise RMS stayed below 8 µV per channel.3
- The central advance is that multiple functionally distinct neural drives were separated within a single reinnervated muscle, removing the need to surgically divide a nerve into fascicles as regenerative peripheral nerve interfaces require.2
- Decoded motor-unit clusters mapped to diverse phantom-limb tasks (for example, index-finger extension and tripod grasp), each repeated four times by one participant and six times by the other two, with real-time visual EMG feedback normalized to task-specific maximum voluntary contraction.23
- About a third of the motor units showed satellite potentials time-locked 8.8–32 ms from the main spike, a waveform signature associated with reinnervation.3
- Published in Nature Biomedical Engineering, the work argues a “hyper-reinnervated muscle” biointerface could supply amputees with more independent prosthetic-control channels and clarify how the central nervous system re-encodes movement after nerve transfer; analyses were offline, so real-time prosthetic control remains future work.234