The role of surface EMG in predicting responsiveness of muscles to FES therapy after cervical SCI.
Guijin Li, Gustavo Balbinot, Sharmini Atputharaj and 4 others
PMID 41204357WHAT IT FOUND
Baseline surface EMG predicted which upper limb muscles would strengthen during FES after cervical SCI better than clinical scores.
The best model caught many non-responders but missed many true responders, so it is not ready for routine use.
Key findings
01In 132 muscles from 17 participants, baseline sEMG predicted FES responder status better than clinical variables alone.
02The best sEMG model had accuracy 0.76, recall 0.42, and true negative rate 0.92, so it identified many non-responders but missed many true responders.
03Clinical variables alone had recall 0.21, lower than the expected 0.33, so many true responders were classified as non-responders.
STILL TO COME
How it was doneWhat they foundWhat it means for PTs
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 analysis used 132 muscles from 17 participants, so it is small for machine learning. The dataset had more non-responders than responders, and the models missed many true responders. There was only one female participant, and the sample overrepresented AIS D and C3-C4 injuries. The models were evaluated only with internal leave-one-participant-out testing, not external validation. Responder status depended on manual muscle test change over therapy, which may not capture all functional recovery. Intramuscular EMG was available for only a subset of muscles and was used as supplementary information.
Declared interests
The paper names Wings for Life as the funder.
The easy way to misread this
Do not conclude that surface EMG can now decide which muscles receive FES. The best model still missed many true responders, and the study was small and imbalanced.
Summarised by AI from the full paper, without a clinician reviewing it. Check it against the source before it changes what you do. Read it on PubMed →