Page 332 - Handbook of Biomechatronics

Upper Extremity Rehabilitation Robots: A Survey 325

3 CLASSIFICATION BY TRAINING

Based on Brewer et al. (2007), these robotic systems can be catego-
rized by training approaches. Accordingly, these robots are classified as either
unilateral or bilateral trainers. Unilateral trainers compromise repetitive
practice of a single arm, while bilateral trainers perform bimanual therapy.
Compared with unilateral trainers, there are a limited number of bilateral
devices available in the literature (Sheng et al., 2016). Both the classes of
trainers can provide gross and/or fine motor movements.
In gross motor movements, massed practice with explicit learning is
accomplished. Gross motor movement is an established method of therapy
used in various rehabilitation robots. Unilateral trainers, such as “MIT-
MANUS” (Krebs et al., 1998), “GENTLE/S” (Loureiro et al., 2003),
“MariBot” (Rosati et al., 2005), “ARM Guide” (Kahn et al., 2006), and
“ARMin” (Nef et al., 2007), and bilateral trainer “MIME” (Burgar et al.,
2000) are used for gross motor movements.
Fine motor movements are mostly related to hand and wrist rehabilita-
tion. This method can be used for increasing ROM or regulation of motor
tasks like independent movements of fingers. Unilateral trainers, such as
“Hand Mentor” (Koeneman et al., 2004), “HEXORR” (Schabowsky
et al., 2010), “HandTutor” (Carmeli et al., 2011), “Amadeo” (Sale et al.,
2012), and “VAEDA glove” (Thielbar et al., 2017), and bilateral trainer
“Bi-Manu-Track” (Hesse et al., 2003b) provide fine motor movements.
Some rehabilitation robots can be used for both gross and fine motor
movements. “RUPERT” (Sugar et al., 2007), the single arm “CADEN-7”
(also known as “EXO-UL7”) (Perry et al., 2007; Simkins et al., 2013),
“ARMin III” (Nef et al., 2009), and “Universal Haptic Drive” (Oblak
et al., 2010) are unilateral trainers of this type. The double arm “EXO-UL7”
(Rosen and Perry, 2007; Simkins et al., 2013) is a bilateral trainer that
provides both gross and fine motor movements.
Together with the above tasks, some robots have additional features such
as real-world practice (Patton et al., 2004), functional electrical stimulation
(FES) (Hu and Tong, 2014), electromyography (EMG) (Rahman et al.,
2015), electroencephalogram (EEG) (Fok et al., 2011), gravity compensa-
tion (Stienen et al., 2007; Moubarak et al., 2010), feedback distortion
(Brewer et al., 2008), telerehabilitation (Ivanova et al., 2015), and progress
assessment (e.g., “KINARM” is used for motor function assessments
(Coderre et al., 2010; Mostafavi et al., 2015)).