A Systematic Review of Machine Learning in Robotics-Assisted Rehabilitation
Abstract
Robotic
technology
is
expected
to
transform
rehabilitation
settings,
by
providing
precise,
repetitive,
and
task-specific
interventions,
thereby
potentially
improving
patients’
clinical
outcomes.
Artificial
intelligence
(AI)
machine
learning
(ML)
have
been
widely
applied
in
different
areas
support
robotic
rehabilitation,
from
controlling
robot
movements
real-time
patient
assessment.
To
provide
overview
the
current
landscape
impact
of
AI/ML
use
robotics
we
performed
a
systematic
review
focusing
on
AI
broad
perspective,
encompassing
pathologies
body
districts,
considering
both
motor
neurocognitive
rehabilitation.
We
searched
Scopus
IEEE
Xplore
databases,
studies
involving
human
participants.
After
article
retrieval,
tagging
phase
was
carried
out
devise
comprehensive
easily-interpretable
taxonomy:
its
categories
include
aim
within
system,
type
algorithms
used,
location
robots
sensors.
The
selected
articles
span
multiple
domains
diverse
aims,
such
as
movement
classification,
trajectory
prediction,
evaluation,
demonstrating
potential
ML
revolutionize
personalized
therapy
improve
engagement.
reported
highly
effective
predicting
intentions,
assessing
outcomes,
detecting
compensatory
movements,
insights
into
future
interventions.
Our
analysis
also
reveals
pitfalls
this
area,
explainability
issues
poor
generalization
ability
when
these
systems
are
real-world
settings.

Research Square (Research Square), Journal Year: 2024, Volume and Issue: unknown
Published: Sept. 3, 2024
Language: Английский