Dynamic Neural Mechanisms Underlying Sustained Motor Training Post-Stroke: An fNIRS Investigation Employing Time-Varying ALFF and Functional Connectivity Analysis.

Abstract

Motor dysfunction of the upper limbs following a stroke predominantly arises from abnormal motor patterning caused by the disrupted balance of inter-cortical communication within motor-associated cortical regions. Temporal analysis offers a more precise reflection of the cortical functional state in affected patients. This study employed fNIRS to capture hemodynamic responses among 20 stroke patients and 19 healthy controls in both resting and Baduanjin task state. Including computing the coefficient of variation of fractional amplitude of low-frequency fluctuations (fALFF) for each channel, alongside extracting dynamic state metrics. Findings indicate that, during sustained motor tasks, stroke patients exhibit a diminished fALFF variability in targeted cortical regions; these individuals display a higher prevalence of low-intensity states and a lower prevalence of high-intensity states during task execution. These dynamic state attributes are significantly associated with scores on the upper limb motor function scale (FMA-UE), thereby proposing a time-domain perspective for investigating the underlying mechanisms of stroke-induced motor dysfunction.