PilotJournal of neuroengineering and rehabilitation2025

Real-time locomotion mode detection in individuals with transfemoral amputation and osseointegration.

Bahareh Ahkami, Morten B Kristoffersen, Max Ortiz-Catalan

PMID 40551132

WHAT IT FOUND

Surface EMG combined with inertial sensors allowed real-time detection of walking, stair, and ramp transitions in five people with transfemoral amputation and osseointegration.

Individual performance varied widely, with error rates ranging from 1.2% to 23%.

Key findings

01Real-time locomotion mode detection was feasible in all five participants, but accuracy varied substantially between individuals.

02Offline classification error was lowest when combining EMG and IMU signals (5%) compared to IMU alone (8%) or EMG alone (25%).

03Prediction delays occurred because the system only made decisions at heel contact and toe-off, potentially delaying recognition by up to half a stride.

STILL TO COME

How it was doneWhat they found

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.

Already have one?

What it does not show

The study included only five participants, all of whom could walk without assistance, limiting generalizability. Participants used passive prosthetic legs, not powered ones, so the interaction between user intent and device resistance was absent. The classifier did not include a standing or resting mode, reducing ecological validity. Real-time predictions were not fed back to the prosthetic, and participants did not receive any feedback during testing. Manual tagging of transitions by a researcher introduced potential timing errors, though some were corrected algorithmically.

Declared interests

Funded by Chalmers University of Technology. No commercial conflicts of interest were declared in the text.

The easy way to misread this

Do not interpret this as evidence that EMG control is ready for clinical use. The technology is not yet ready for widespread deployment in take-home devices due to significant delays in transition detection and variable signal quality during movement.

Read it on PubMed →


The study

Participants
5 participants
Certainty of evidence
Very low

Browse

    Cite

    Bahareh Ahkami, Morten B Kristoffersen, Max Ortiz-Catalan Real-time locomotion mode detection in individuals with transfemoral amputation and osseointegration. Journal of neuroengineering and rehabilitation. 2025.

    Read the original