International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering,
Journal Year:
2024,
Volume and Issue:
14(2), P. 2100 - 2100
Published: Jan. 26, 2024
Ataxic
gait
monitoring
and
assessment
of
neurological
disorders
belong
to
important
areas
that
are
supported
by
digital
signal
processing
methods
artificial
intelligence
(AI)
techniques
such
as
machine
learning
(ML)
deep
(DL)
techniques.
This
paper
uses
spatio-temporal
data
from
Kinect
sensor
optimize
model
distinguish
between
ataxic
normal
gait.
Existing
ML-based
methodologies
fails
establish
feature
correlation
different
parameters;
thus,
exhibit
very
poor
performance.
Further,
when
is
imbalanced
in
nature
the
existing
induces
higher
false
positive.
In
addressing
research
issues
this
introduces
an
extreme
gradient
boost
(XGBoost)-based
classifier
enhanced
optimization
(EFO)
modifying
standard
cross
validation
(SCV)
mechanism.
Experiment
outcome
shows
proposed
person
identification
achieves
good
result
comparison
with
DL-based
methodologies.
IEEE Transactions on Neural Systems and Rehabilitation Engineering,
Journal Year:
2024,
Volume and Issue:
32, P. 412 - 421
Published: Jan. 1, 2024
Gait
impairment
in
Parkinson's
Disease
(PD)
is
quantitatively
assessed
using
the
Movement
Disorder
Society
Unified
Rating
Scale
(MDS-UPDRS),
a
well-established
clinical
tool.
Objective
and
efficient
PD
gait
assessment
crucial
for
developing
interventions
to
slow
or
halt
its
advancement.
Skeleton-based
MDS-UPDRS
score
estimation
has
attracted
increasing
interest
improving
diagnostic
efficiency
objectivity.
However,
previous
works
ignore
important
cross-spacetime
dependencies
between
joints
gait.
Moreover,
existing
skeleton
datasets
are
very
small,
which
big
issue
deep
learning-based
studies.
In
this
work,
we
collect
sizable
dataset
by
multi-view
Azure
Kinect
sensors.
The
collected
contains
102
patients
30
healthy
older
adults.
addition,
data
from
16
young
adults
(aged
24-50
years)
further
examine
effect
of
age
on
assessment.
For
skeleton-based
automatic
analysis,
propose
novel
cross-spatiotemporal
graph
convolution
network
(CST-GCN)
learn
complex
features
patterns.
Specifically,
labeling
strategy
designed
assemble
group
neighbors
root
node
according
spatiotemporal
semantics
skeleton.
Based
strategy,
CST-GCN
module
explicitly
models
among
joints.
Finally,
dual-path
model
presented
realize
modeling
fusion
spatial,
temporal,
features.
Extensive
experiments
validate
effectiveness
our
method
dataset.
Sensors,
Journal Year:
2023,
Volume and Issue:
23(8), P. 3902 - 3902
Published: April 12, 2023
Parkinson's
disease
(PD)
is
characterized
by
a
variety
of
motor
and
non-motor
symptoms,
some
them
pertaining
to
gait
balance.
The
use
sensors
for
the
monitoring
patients'
mobility
extraction
parameters,
has
emerged
as
an
objective
method
assessing
efficacy
their
treatment
progression
disease.
To
that
end,
two
popular
solutions
are
pressure
insoles
body-worn
IMU-based
devices,
which
have
been
used
precise,
continuous,
remote,
passive
assessment.
In
this
work,
insole
were
evaluated
impairment,
subsequently
compared,
producing
evidence
support
instrumentation
in
everyday
clinical
practice.
evaluation
was
conducted
using
datasets,
generated
during
study,
patients
with
PD
wore,
simultaneously,
pair
instrumented
set
wearable
devices.
data
from
study
extract
compare
features,
independently,
aforementioned
systems.
Subsequently,
subsets
comprised
extracted
machine
learning
algorithms
impairment
results
indicated
kinematic
features
highly
correlated
those
Moreover,
both
had
capacity
train
accurate
models
detection
impairment.
Expert Review of Neurotherapeutics,
Journal Year:
2023,
Volume and Issue:
23(8), P. 689 - 702
Published: June 27, 2023
Although
clinician-based
assessment
through
standardized
clinical
rating
scales
is
currently
the
gold
standard
for
quantifying
motor
impairment
in
Parkinson's
disease
(PD),
it
not
without
limitations,
including
intra-
and
inter-rater
variability
a
degree
of
approximation.
There
increasing
evidence
supporting
use
objective
motion
analyses
to
complement
assessment.
Objective
measurement
tools
hold
significant
potential
improving
accuracy
research-based
evaluations
patients.The
authors
provide
several
examples
from
literature
demonstrating
how
different
tools,
optoelectronics,
contactless
wearable
systems
allow
both
quantification
monitoring
key
symptoms
(such
as
bradykinesia,
rigidity,
tremor,
gait
disturbances),
identification
fluctuations
PD
patients.
Furthermore,
they
discuss
how,
clinician's
perspective,
measurements
can
help
various
stages
management.In
our
opinion,
sufficient
supports
assertion
that
enable
accurate
evaluation
complications
PD.
A
range
devices
be
utilized
only
support
diagnosis
but
also
monitor
symptom
during
progression
become
relevant
therapeutic
decision-making
process.
npj Digital Medicine,
Journal Year:
2024,
Volume and Issue:
7(1)
Published: June 26, 2024
Abstract
Gait
impairments
are
among
the
most
common
and
disabling
symptoms
of
Parkinson’s
disease
worsen
as
progresses.
Early
detection
diagnosis
subtype-specific
gait
deficits,
well
progression
monitoring,
can
help
to
implement
effective
preventive
personalized
treatment
for
PD
patients.
Yet,
features
have
not
been
fully
studied
in
its
motor
subtypes.
To
characterize
comprehensive
objective
alterations
identify
potential
biomarkers
early
diagnosis,
subtype
differentiation,
severity
monitoring.
We
analyzed
parameters
related
upper/lower
limbs,
trunk
lumbar,
postural
transitions
from
24
tremor-dominant
(TD)
20
instability
difficulty
(PIGD)
dominant
patients
who
were
stage
39
matched
healthy
controls
(HC)
during
Timed
Up
Go
test
using
wearable
sensors.
Results
show:
(1)
Both
TD
PIGD
groups
showed
restricted
backswing
range
bilateral
lower
extremities
more
affected
side
(MAS)
arm,
reduced
lumbar
rotation
coronal
plane,
low
turning
efficiency.
The
receiver
operating
characteristic
(ROC)
analysis
revealed
these
had
high
discriminative
value
distinguishing
both
subtypes
HC
with
area
under
curve
(AUC)
values
0.7~0.9
(
p
<
0.01).
(2)
Subtle
but
measurable
differences
existed
between
before
onset
clinically
apparent
impairment.
(3)
Specific
significantly
associated
Objective
based
on
sensors
may
facilitate
timely
treatments
through
International Journal of Social Robotics,
Journal Year:
2025,
Volume and Issue:
unknown
Published: Jan. 9, 2025
Abstract
In
standard
clinical
protocols,
the
result
of
neuromotor
rehabilitation
programs
is
evaluated
through
validated
scales
and
tests
able
to
measure
motor
performance
patients
monitor
their
improvements
over
time.
The
Timed
Up
Go
(TUG)
test
one
most
common
assessments
used
evaluate
patients’
dynamic
balance,
as
well
mobility.
However,
in
its
traditional
version,
TUG
does
not
provide
quantitative
information
on
gait
performances—only
subjectively
observed
by
clinician—and
timing
different
phases
involved
execution.
availability
additional
would
indeed
be
useful
for
clinicians
formulate
a
more
accurate
assessment
patient
define
personalized
treatment
plan.
this
sense,
use
Socially
Assistive
Robots
(SARs)
could
help
improving
performance,
relieving
at
same
time
physiotherapists
from
consuming
tasks.
goal
feasibility
study
twofold:
(1)
assess
quality
functionality
implemented
robot
technical
standpoint
(2)
perception
“R1-TUG”
solution
potential
end-users
point
view,
terms
usability
acceptability.
A
set
involving
sample
healthy
volunteers
revealed
that
adoption
SAR
an
tool,
improve
ability
physiotherapist
objectively
subject’s
movement
while
ensuring
adequate
level
acceptability
participants.
This
work
represents
promising
future
robotic
solutions
within
context.
