IEEE Robotics & Automation Magazine - March 2023 - 70

module is a bit bulky, participants felt comfortable and confident
using the system.
During the stage of developing AA, the therapists proposed
some changes in the training games, such as 1) substituting
the image of a cup for a regular glass in the At the
Bar game in order not to promote the consumption of alcohol
or 2) increasing the number of puzzle pictures available to
increase participant engagement. On the other hand, as the
participants did not have any cognitive impairment, some
games were considered too simple and easy, and therefore,
boring. As a result, proposed changes included increasing the
variety and complexity of training games that demand higher
cognitive loads.
DISCUSSION
While the functionality and usability of AA have been well
accepted by therapists and patients through clinical trials,
some recommendations have been proposed. In the following
text, some technical improvements are discussed.
First, it has been noted that the AA wheels can smear the
mat due to friction when the AA system is used for extended
periods, resulting in an additional delay in the absolute
position and orientation update because the DM images are
not sufficiently clear. Hence, the mat must be periodically
cleaned with a wet wipe by the user. After long periods of
time, it may be necessary to replace the mat or wheels that
get damaged or dirty with long-term use. Second, as previously
stated, the position estimation of the arm in the plane
is based on DM images detected, which implies the use of a
camera and the mat. The use of a camera for position detection
requires a somewhat bulky dome to read the DM properly
due to the distance needed between the camera and the
TABLE 4. Usability evaluation: Demographic data of
participants and results.
DEMOGRAPHIC DATA
Age (years, mean ± SD)
Sex (male/female)
Side hemiparesis (left/right)
Stage (period since stroke in months)
Subacute (2-6)
Chronic of short evolution (6-12)
Long-term chronic (>12)
USABILITY TEST RESULTS
SUS (0-100)
QUEST (1-5)
IMI (1-7)*
Interest
Usefulness
THE NETHERLANDS
64.75 ± 8.52
8/4
7/5
12
77.27 ± 16.1
3.90 ± 0.39
5.5 ± 0.5
6.5 ± 0.17
SPAIN
63.89 ± 15.37
6/3
6/3
3
3
3
71.94 ± 16.38
3.81 ± 0.38
6.0 ± 1.82
6.1 ± 1.93
*The results of the IMI in The Netherlands were calculated with N = 10 because one of the participants
was unable to complete the questionnaire. SD: standard deviation; SUS: System Usability
Scale; IMI: Intrinsic Motivation Inventory; QUEST: Quebec User Evaluation of Satisfaction with
Assistive Technology.
mat. However, this technology was selected for its cost-effectiveness.
Changing the technology used for the calculation of
the position in the plane could result in a smaller device and
remove the mat component and/or the delay existing in the
DM interpretation.
A third observation was made during preliminary tests with
3D-printed plastic parts, where the wrist and hand modules
were broken during the trials with real subjects as a result of
strong grasping forces. So, more rigid and robust pieces were
manufactured for the usability test presented in this article.
In terms of the telerehabilitation platform, therapists
proposed some improvements in the training games. Specifically,
they proposed making them more sophisticated by
including 3D images or virtual reality systems to engage the
subjects.
The inclusion of the wrist and hand modules was highly
valued for the rehabilitation of prono-supination and hand
grasping/opening. Nevertheless, the rehabilitation of wrist
flexion/extension has also been required by some clinicians,
so a new design of the module could be developed to include
these movements.
In a new and enhanced version, the functionality of AA
has been extended by including motors in the mobile base. In
this way, AA can provide active assistance that is necessary
to rehabilitate subjects with severe impairment in the acute
phase who cannot otherwise perform the active shoulder and
elbow movements needed for rehabilitation. The presented
nonmotorized module is deployed for home use, while
a motorized module has been proposed for rehabilitation
facilities under clinicians' supervision to ensure safety while
allowing consistent rehabilitation throughout the entire period
of therapy.
CONCLUSION
AA is an at-home rehabilitation solution
that includes a robot that measures
the active movement of the arm,
wrist, and hand grasping and is supported
by a telerehabilitation platform
based on serious games for assessment
and long-term training. AA consists
of a portable robotic module and
dedicated software that can be easily
installed on any PC or tablet to be
used at home. AA has been tested by
real users, demonstrating both clinical
effectiveness and acceptance of the
system, with several minor recommendations
for improvement.
Recently, the advancement of information
and communication technology
(ICT), such as the Internet of Things,
data science, and artificial intelligence,
has allowed innovative changes in the
industrial and automotive sectors and
has finally reached the medical field.
70 IEEE ROBOTICS & AUTOMATION MAGAZINE MARCH 2023

IEEE Robotics & Automation Magazine - March 2023

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - March 2023