Simplification of Mobility Tests and Data Processing to Increase Applicability of Wearable Sensors as Diagnostic Tools for Parkinson’s Disease
Sensors,
Год журнала:
2024,
Номер
24(15), С. 4983 - 4983
Опубликована: Авг. 1, 2024
Quantitative
mobility
analysis
using
wearable
sensors,
while
promising
as
a
diagnostic
tool
for
Parkinson's
disease
(PD),
is
not
commonly
applied
in
clinical
settings.
Major
obstacles
include
uncertainty
regarding
the
best
protocol
instrumented
testing
and
subsequent
data
processing,
well
added
workload
complexity
of
this
multi-step
process.
To
simplify
sensor-based
diagnosing
PD,
we
analyzed
from
262
PD
participants
50
controls
performing
several
motor
tasks
wearing
sensor
on
their
lower
back
containing
triaxial
accelerometer
gyroscope.
Using
ensembles
heterogeneous
machine
learning
models
incorporating
range
classifiers
trained
set
features,
show
that
our
effectively
differentiate
between
with
controls,
both
mixed-stage
(92.6%
accuracy)
group
selected
mild
only
(89.4%
accuracy).
Omitting
algorithmic
segmentation
complex
decreased
accuracy
models,
did
inclusion
kinesiological
features.
Feature
importance
revealed
Timed
Up
Go
(TUG)
to
contribute
highest-yield
predictive
minor
decreases
based
cognitive
TUG
single
task.
Our
approach
facilitates
major
simplification
without
compromising
performance.
Язык: Английский
Inertial-based dual-task gait normalcy index at turns: a potential novel gait biomarker for early-stage Parkinson’s disease
IEEE Transactions on Neural Systems and Rehabilitation Engineering,
Год журнала:
2025,
Номер
33, С. 687 - 695
Опубликована: Янв. 1, 2025
As
one
of
the
main
motor
indicators
Parkinson's
disease
(PD),
postural
instability
and
gait
disorder
(PIGD)
might
manifest
in
various
but
subtle
symptoms
at
early
stage
resulting
relatively
high
misdiagnosis
rate.
Quantitative
assessment
under
dual
task
or
complex
(i.e.,
turning)
may
contribute
to
better
understanding
PIGD
provide
a
diagnostic
indicator
early-stage
PD.
However,
few
studies
have
explored
deviation
evaluation
algorithms
that
reflect
specificity.
In
this
work,
we
proposed
novel
inertial-based
normalcy
index
(GNI)
based
on
quantitative
model
characterize
overall
performance
during
both
straight
walking
turning
with
without
serial-3
subtraction
task.
The
factor
group
GNI
variable
was
investigated
feasibility
improve
PD
validated.
experimental
results
showed
paradigm
is
significant
where
dual-task
turn
had
best
discriminating
ability
between
HC
(AUC
=
0.992)
significantly
correlated
UPDRS
III
(r
0.81),
MMSE(r
0.57)
Mini-BEST(r
0.65).
We
also
observed
turning-based
has
larger
effect
size
compared
clinical
scales,
demonstrating
can
changes
functional
mobility
rehabilitation
for
Our
work
offers
an
innovative
potential
biomarker
diagnostics
provides
new
perspective
into
its
application
Язык: Английский