EEG based multifunctional connectivity fusion across frequency bands and parameters promote motor function assessment in stroke: a pilot study.

The study included only 12 stroke patients and 12 healthy controls, so it has low statistical power and limited generalizability. Many patients had mild upper limb impairment and near-ceiling FMA-UE scores, while only a small number had severe deficits. No correction for multiple comparisons was applied because it was a pilot, so the findings are exploratory. Fusion weights and best correlations were optimized on the same dataset, and the best result was selected from multiple fusion strategies, which may overstate performance. The study used only resting-state EEG and did not test task-related or longitudinal changes during motor recovery. All participants were right-handed, so the results do not show how the method behaves in left-handed people.

Funding was reported from the National Natural Science Foundation of China and the Tianjin Science and Technology Program - State Key Laboratory Major Special Project. The supplied text does not report author conflicts of interest.

Do not read the strongest upper limb result as proof that this EEG method can guide stroke rehabilitation. The best fusion value was selected from many combinations on the same dataset of 12 stroke patients, and the study was exploratory.