Biomedical Engineering Reference
In-Depth Information
here some concepts that should be considered for the design and implementation
of exercises for hand rehabilitation.
Passive and active exercises are currently used in robot-assisted rehabilitation
(Marchal-Crespo and Reinkensmeyer (2009)), with robots producing assistive or
resistive forces, or more sophisticated assist-as-needed strategies, or event related
assistance based on brain or muscle activity. Typically, at an early stage in stroke
recovery, movements such as finger extension can only be performed passively,
i.e., the robot has to move the fingers in order to compensate for the weakness
in finger extensors observed in stroke subjects. In this phase, it is crucial that the
robot monitors the interaction forces and gently moves the fingers.
However, passive movements driven by robotic interfaces may not be suf-
ficient for hand rehabilitation; while passive movements contribute to reducing
muscle tone and maintaining passive properties of joints and muscles (Hesse
et al.
(2003)), active movements initiated by the subject are required to promote correct
patterns of muscle activation and improve strength (Hogan
et al.
(2006), Woldag
et al.
(2007)), which are crucial to dexterous manipulation.
Furthermore, active participation of the subject during training is fundamental
to skill acquisition (Winstein
et al.
(2003)). Motor recovery after stroke is consid-
ered as a form of motor learning, where undamaged brain regions are recruited
to generate motor commands to the same muscles that were used before the
injury. It may, therefore, be advantageous to apply the motor learning principles
observed in healthy subjects to improve stroke rehabilitation. Note that active
participation can also be achieved during passive exercises by giving subjects
proper instructions, but movements remain limited by the passive guidance and
are thus far from real activities. In this sense, therapy should focus on intensive
repetition of active movements where subjects interact with various force fields to
help develop strategies that will be optimal for improving hand function in ADL
(Krakauer (2006), Reinkensmeyer
et al.
(2004)).
However, intensive repetition of movements involving the impaired limb
require considerable effort by the subject and may be painful and tiring. Ad-
ditionally, after several repetitions, subjects may get bored, leading to decreased
concentration and motivation to train, thus affecting the quality of the therapy.
Motivation is important to ensure that subjects are engaged and train at
their full potential during therapy sessions, pushing the limits imposed by their
impairment. A common way to keep subjects motivated during intense movement
repetitions in the use of interactive feedback. Visual feedback is widely used in
robot-assisted rehabilitation systems, and is probably the simplest way to interact
with a user. Moreover, exercises with visual feedback or virtual environments may
help improve the outcome of post-stroke rehabilitation, by augmenting subjects'
awareness of their actual performance, and promoting visuomotor coordination
(Henderson
et al.
(2007), Holden (2005)).
Robotic devices have the advantage of providing haptic feedback, i.e. specific
force and/or torque patterns that can stimulate proprioceptive sensors in skin,
joints and muscles to increase subjects' awareness and use of somatosensory
