Resting-state and task-state fNIRS fusion assessment method for upper limb motor function in stroke patients.
Ziwen Yuan, Weiwei Xu, Zhehao Hu and 8 others
PMID 41913241WHAT IT FOUND
Combining resting and hand-grasp fNIRS measures with clinical data predicted upper-limb Fugl-Meyer scores better than either fNIRS measure alone.
Cortical stroke predictions were closest. Subcortical stroke predictions remained less accurate.
Key findings
01In cortical stroke, the resting-state plus task-state fNIRS fusion model had the best reported fit, with mean absolute error 5.11 and 11 of 19 patients within 5 points at 45% sparsity.
02In subcortical stroke, the fusion model also outperformed single-state models, but errors were larger than in cortical stroke, with mean absolute error 7.39 at 35% sparsity.
03Fusion of clinical variables with resting-state and task-state fNIRS fit better than models using clinical variables alone or fNIRS alone.
STILL TO COME
How it was doneWhat they foundWhat it means for PTsWhat 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
The study used only 57 patients, and the authors state that statistical power and overfitting risk need careful evaluation. The models were validated only within the same sample using leave-one-out testing, not in an independent group. Patients with moderate FMA-UL scores were underrepresented, so predictions across the full severity range are uncertain. Subcortical stroke predictions were less accurate than cortical stroke predictions, especially for scores in the 25 to 40 FMA-UL range. Some patients could not physically grip and used motor imagery, which may confound task-state connectivity. Patients with significant aphasia, dysphrenia, or cognitive impairment were excluded, so the method was not tested in those groups. The paper does not test whether using the model changes therapy or patient outcomes.
Declared interests
Funded by the National Natural Science Foundation of China, the Key Technologies Research and Development Program, and the Key Research and Development Program of Shaanxi. The supplied text gives no competing-interest statements.
The easy way to misread this
Do not treat this as a ready-to-use bedside test. It was built on 57 patients, used internal validation only, and predicted subcortical stroke scores less accurately. The reported error numbers describe model prediction, not proven clinical benefit.
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 →