Therapeutics
Mengjian Qu, MS
Chief therapist
Rehabilitation department
Hengyang City, Hunan, China (People's Republic)
Huali Tang, MS
None
Department of Rehabilitation ，Rehabilitation Medicine Center ，Rehabilitation Laboratory， the First Affiliated Hospital ，Hengyang Medical School， University of South China
Hengyang, Hunan, China (People's Republic)
Jun Zhou, PhD
head of department
Rehabilitation department
Hengyang City, Hunan, China (People's Republic)
This study aimed to characterize and compare the distinct neural strategies elicited by different training modes in stratified subgroups of stroke patients. We tested the novel hypothesis that 'mode sensitivity'—the degree to which cortical activation is modulated by changes in therapy mode—serves as a biomarker for the brain's recovery state.
Design:
Forty-one stroke patients, stratified by functional level, chronicity, and hemiplegic side, performed tasks in four distinct modes (Passive, Assistive-Active, Active, Mirror) delivered by a robotic exoskeleton. A 48-channel fNIRS system monitored cortical activation. 'Mode sensitivity' was assessed via FDR-corrected, intra-group comparisons of activation across modes.
Results:
A primary and crucial finding was the absence of any statistically significant differences in global mean activation intensity between any of the stratified subgroups during the active mode (p > 0.05), indicating that overall neural effort was comparable. The core distinction was instead revealed in the activation strategy. The low-function, subacute, and left-hemiplegia subgroups were all highly "mode-sensitive." They exhibited significant and widespread modulation of activation across key motor cortices (M1, PMC) in response to the specific demands of each mode. Critically, significant activation differences were observed not only between passive and active states but also among the different active-participation modes (e.g., Active vs. Mirror, p < 0.05). Conversely, the high-function, chronic, and right-hemiplegia subgroups were "mode-consolidated." They demonstrated a stable, fixed activation pattern with almost no statistically significant differences in cortical activation among the Assistive-Active, Active, and Mirror modes.
Conclusion:
Post-stroke neural recovery is characterized by a strategic evolution from a flexible, cue-dependent "mode-sensitive" state to a more automated "mode-consolidated" state. This distinction provides a robust neurophysiological rationale for personalizing rehabilitation, enabling clinicians to strategically match the therapeutic mode to a patient's specific neural profile, rather than defaulting to a single approach.
