Biomedical Engineering Reference
In-Depth Information
In a robot-assisted CPM treatment, there is no voluntary effort required from
a patient, and the movements of the patient's limb(s) on the paretic side are
dominated by the robot system. This type of intervention was found to be effective
in temporarily reducing poststroke hypertonia (Schmit
et al
. 2000). In active-
assisted robotic treatment (or interactive robotic treatment), the rehabilitation robot
would provide external assisting forces to compensate the weakness in the affected
limb when a patient could not complete a desired movement independently. It has
been found that with the involvement of voluntary motor efforts of the paretic
limbs, the limb functional improvement after interactive robotic treatments were
better than those achieved by the CPM treatments (Volpe
et al
. 2004; Tong
et
al.
2006). In a robot-assisted treatment with challenge-based moments, the robot
controller will make a training task more difficult or challenging, as opposed to the
robotic assistance provided in an interactive robotic treatment. For example, the
commonly used challenging control is robotic resistance (Fasoli
et al
. 2003). It has
been found that repetitive practice of hand and finger movements against resistive
loads resulted in greater improvements in motor performance and functional
scales than Bobath-based treatment (Butefisch
et al
. 1995), transcutaneous electric
nerve stimulation, and suprathreshold electric stimulation on hand and wrist
muscles (Hummelsheim
et al
. 1997).
In recent years, robotic systems developed for stroke rehabilitation have been
widely reported in the literatures. Most of these robotic systems can provide all the
three motion types mentioned above for poststroke rehabilitation. For example,
MIT-MANUS is one of the robotic systems for poststroke upper limb rehabilitation
(Hogan
et al
. 1992; Krebs
et al
. 1998). The key feature of MIT-MANUS is its
impedance control which could keep a compliant trajectory under perturbation.
ARM Guide is a robotic system designed both for training and evaluation of
upper limb reaching functions in a linear trajectory (Reinkensmeyer
et al.
1999;
Reinkensmeyer
et al.
2000). Colombo
et al
. also designed a wrist manipulator with
one-degree of freedom and an elbow-shoulder manipulator with two degrees of
freedom for the rehabilitation of upper limb movements (Colombo
et al
. 2005).
They used admittance control to reduce the inertia and facilitate the movement.
Besides, other rehabilitation robots with different mechanical designs also have
been well described in other chapters for hand and upper limb trainings (
Chapter
2
and
4
).
Due to the believed effectiveness in motor improvement by active-assisted
robotic treatment (i.e., in comparison with the CPM training mode), the recent
developments involving rehabilitation robots has been worked towards the inter-
active control, which allows the robotic system to react to patients' voluntary inten-
tion (Marchal-crespo and Reinkensmeyer 2009). However, most of the interactive
controls used in rehabilitation robots are oversimplified on the voluntary inputs
from stroke patients and lacking clinical supports on their training effectiveness.
The easiest and most commonly used algorithm for the interactive control is the
“on/off” strategy, i.e., once the voluntary indicator from a subject is captured or
disappears, the assistive function of robotic system will be triggered to help the
