Decoding the grasping intention from electromyography during reaching motions.
Iason Batzianoulis, Nili E Krausz, Ann M Simon and 2 others
PMID 29940991WHAT IT FOUND
EMG during reaching predicted grasp types better in later phases than early.
The tested algorithms did not clearly differ. In one real-time test, all-phase training was faster and more accurate than final-phase training.
Key findings
01EMG activity differed significantly across the three reach-to-grasp phases for all subjects.
02Offline grasp-type accuracy was poor in the first phase and improved in the second and third phases for both amputee and able-bodied subjects.
03In one amputee's real-time test, training on all motion phases gave 80±5% accuracy and reached confidence in 0.3±0.10 s, compared with 55±5% accuracy and 0.42±0.12 s when trained only on the third phase.
STILL TO COME
How it was doneWhat they foundWhat 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
Only four transradial amputees participated, and the real-time prosthesis test involved one amputee, so the findings are a proof of concept rather than clinical evidence. The paper reports a discrepancy in participant count: the experimental protocol says eight able-bodied subjects, while the conclusion says four able-bodied subjects. The task used five predefined grasp types and a single wrist orientation, so it does not test everyday objects, different orientations, or wrist control. The reported outcomes are classification accuracy and prediction time, not patient function, comfort, prosthesis rejection, or quality of life. TMR versus non-TMR comparisons involved two subjects per group and the authors state they should be taken with caution. The real-time device was a prototype RIC hand with two degrees of actuation, not a commercial prosthesis.
Declared interests
The supplied text does not include an author conflict-of-interest declaration. The article is listed as supported by NIH extramural and non-U.S. government research support.
The easy way to misread this
Do not conclude that this method can reliably decode grasp intention for prosthetic users in practice. The offline accuracy was poor in the first phase, classifier choice did not significantly differ, and the real-time result came from one amputee.