Page 333 - Handbook of Biomechatronics

326 Borna Ghannadi et al.

As discussed in Section 1, there is no significant advantage that can make
one method of training superior to the other. Both unilateral and bilateral
trainers are pursuing the same goal, and their selection depends on the
patient’s condition and his/her level of disability. Hence, plotting a general
guideline for the selection of a suitable trainer is a complicated and cumber-
some procedure, and it is case-dependent. For example, in early stages of
stroke, a unilateral trainer who provides gross movements is a generally
preferable choice. In the next stages, this training can be combined with
real-world practice. For fine movements, if exoskeletons are not affordable,
FES can be used instead. Finally, to quantify functional activities of the sub-
ject, bio-feedback features (EMG and EEG) can be used.

4 CLASSIFICATION BY FORM OF REHABILITATION

Upper extremity rehabilitation robots can support daily activities and
are designed for home or clinical use (Maciejasz et al., 2014). The target pop-
ulation for most of these robotic systems is poststroke patients, for whom
these robots can be active, passive, haptic, or coaching devices.
Active devices provide active/passive assistance therapy. In passive mode,
the robot moves the patient’s limb without any muscular activity of the pas-
sive patient, while in active mode the patient is active during training. Most
upper extremity rehabilitation robots are active devices (Maciejasz et al.,
2014). In contrast to active devices, passive devices perform passive resistance
therapy. These devices are used to provide different types of muscle
strengthening exercises including isometric, isotonic, isokinetic, and iso-
contractile. “Biodex System 4 Pro” is used for isokinetic exercises
(Cvjetkovic et al., 2015), “MEM-MRB” is an isokinetic and iso-contractile
exercise machine (Oda et al., 2009), and “PLEMO” (Kikuchi et al., 2007)
and “WOTAS” (Rocon et al., 2007) are other examples of passive devices.
In addition to active and passive devices, there are some devices that do
not explicitly assist or resist the patient’s movement; instead they are used for
real-world practice. Haptic devices transfer tactile sensing to the patient. They
do not assist or resist movement, but they provide real-world practice by
incorporating haptic feedback while a patient is manipulating virtual objects
in the simulated environment (i.e., virtual reality). There are various exam-
ples of virtual reality in rehabilitation research in which actuated feedback is
implemented (Todorov et al., 1997; Prisco et al., 1998; Jack et al., 2001;
Sveistrup, 2004).In Johnson et al. (2004) and Wamsley et al. (2017), gaming
steering wheels are used to generate force feedback for poststroke upper