IEEE Robotics & Automation Magazine - June 2019 - 31

for participation in this study, a set of inclusion and exclusion criteria was established.
The inclusion criteria are as follows:
● patients 3-10 years of age who suffer from OBPP or ICP
● recruited from or clinically assessed in the Infantile Rehabilitation Service of VRUH
● clinically stable and capable of starting the treatment
● authorization by parents or guardians with the corresponding signed agreement.
The exclusion criteria are as follows.
● visual difficulties
● pain that makes it impossible to perform exercises
● other associated neurological pathologies.
It is important to note that, because each patient presents different conditions and capacities, it is an indispensable requirement of the platform to offer customization for
each patient of the study. This implies that each session,
exercise, and group of postures must be adapted and personalized for each subject and modified, if necessary,
throughout the study.
Experimental Setup
All sessions during the study were performed under the same
conditions. The patients attended all sessions accompanied by
their relatives or caregivers. As shown in Figure 5, the therapist
welcomed them and invited the patient to enter the training
room, while the relatives watched the session from the observation room through a one-way mirror. The patient stood
approximately 1.5 m from the robot, which "feigned being
asleep." The RGB-D sensor was located just behind the robot.
The therapist was seated in front of several laptops to configure and, if necessary, give instructions to the patient. A video
camera was positioned to continuously film each session.
Study Protocol
The patients who participated in this study are very different
in terms of their pathology; furthermore, patients could perform different complementary activities the same as any child
(swimming, dancing, soccer, and so on), which may have
influenced the study's results. In this sense, a randomized,
controlled trial was not possible because the patients could
not be recruited and divided into two groups with similar features. Moreover, the recruitment was difficult, and the number of patients was small; therefore, this article proposes a
quasi-experimental small-N design [41].
The small-N design involves serial observations of single individuals or small groups before, during, and after an
intervention period. It allows researchers to provide clinicians with practical information for making decisions to
improve the care of individual patients. It also offers
potential avenues for including evaluation and research
design in clinical practice and building a foundation for
evidence-based rehabilitation at the individual patient
level in actual treatment settings. Figure 6 shows the
design of the study and the prepost evaluation periods utilized in the assessment.

Prephase
During this phase, patients followed traditional motor training
approved by VRUH. To that end, the rehabilitation physicians
evaluated the patients and
defined objectives based
on patient condition. With
PELEA is a planning and
these objectives, the physiotherapist designed the
replanning system that
personalized training plan
consisting of motor exerwraps an automated
cise repetitions that stimulate and reinforce therapy
planner to provide the
aspects indicated by the
physicians. Rehabilitation
next coherent action
sessions were scheduled
for patients at the hospital
with respect to the
twice per week. On average, the sessions lasted
perceived state.
30 min, and the training
was performed for two
months. The physicians evaluated the patient at the beginning and end of
the training.
Postphase
Because the patients in both phases were the same, the final
assessment of the prephase was considered the initial assessment of the postphase, as shown in Figure 6. During this evaluation, patients performed motor rehabilitation with the
SAR-based platform under the supervision of the physiotherapists. For each session, the system had a set of personalized
motor exercises using the two activities discussed in the
"Autonomous Decision Making" section; depending on the
patient's progress, the training automatically became either
more difficult or more relaxed. The evaluation with the robot
was performed three times: during the first evaluation of the
system (two days after training with the robotic platform),
during the month of training, and at the end of the training
(after two months). Following that, the physicians assessed
the patients for the last time.
Study Variables
Based on the objectives of this study, several aspects were
addressed such as input and output variables, the former corresponding to the patient's data and the latter related to clinical

Patient

Two Months

Prephase
(Traditional)

Postphase
(SAR-Based)

Clinical Outcomes
Questionnaire About the Perception of the Technology
Figure 6. The pre-post evaluation design.

JUNE 2019

IEEE ROBOTICS & AUTOMATION MAGAZINE

IEEE Robotics & Automation Magazine - June 2019

Table of Contents for the Digital Edition of IEEE Robotics & Automation Magazine - June 2019