Adapting to the Mechanical Properties and Active Force of an Exoskeleton by Altering Muscle Synergies in Chronic Stroke Survivors.

2020 
Chronic stroke survivors often suffer from gait impairment resistant to intervention. Recent rehabilitation strategies based on gait training with powered exoskeleton appear promising, but whether chronic survivors may benefit from them remains controversial. We evaluated the potential of exoskeletal gait training in restoring normal motor outputs in chronic survivors (N=10) by recording electromyographic signals (EMGs, 28 muscles both legs) as they adapted to exoskeletal perturbations, and examined whether any EMG alterations after adaptation were underpinned by closer-to-normal muscle synergies. A unilateral ankle-foot orthosis that produced dorsiflexor torque on the paretic leg during swing was tested. Over a single session, subjects walked overground without exoskeleton (FREE), then with the unpowered exoskeleton (OFF), and finally with the powered exoskeleton (ON). Muscle synergies were identified from EMGs using non-negative matrix factorization. During adaptation to OFF, some paretic-side synergies became more dissimilar to their nonparetic-side counterparts. During adaptation to ON, in half of the subjects some paretic-side synergies became closer to their nonparetic references relative to their similarity at FREE as these paretic-side synergies became sparser in muscle components. Across subjects, level of inter-side similarity increase correlated negatively with the degree of gait temporal asymmetry at FREE. Our results demonstrate the possibility that for some survivors, exoskeletal training may promote closer-to-normal muscle synergies. But to fully achieve this, the active force must trigger adaptive processes that offset any undesired synergy changes arising from adaptation to the device's mechanical properties while also fostering the reemergence of the normal synergies.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    51
    References
    4
    Citations
    NaN
    KQI
    []