Article

Many robotic devices in rehabilitation incorporate an assist-as-needed haptic
guidance paradigm to promote training. This error reduction model, while
beneficial for skill acquisition, could be detrimental for long-term retention.
Error augmentation (EA) models have been explored as alternatives. A robotic
Tethered Pelvic Assist Device has been developed to study force application to
the pelvis on gait and was used here to induce weight shift onto the paretic
(error reduction) or nonparetic (error augmentation) limb during treadmill
training. The purpose of these case reports is to examine effects of training
with these two paradigms to reduce load force asymmetry during gait in two
individuals after stroke (>6 mos). Participants presented with baseline gait
asymmetry, although independent community ambulators. Participants underwent 1-hr
trainings for 3 days using either the error reduction or error augmentation
model. Outcomes included the Borg rating of perceived exertion scale for
treatment tolerance and measures of force and stance symmetry. Both participants
tolerated training. Force symmetry (measured on treadmill) improved from
pretraining to posttraining (36.58% and 14.64% gains), however, with limited
transfer to overground gait measures (stance symmetry gains of 9.74% and 16.21%).
Training with the Tethered Pelvic Assist Device device proved feasible to improve
force symmetry on the treadmill irrespective of training model. Future work
should consider methods to increase transfer to overground gait.

Langue : ANGLAIS