Detecting cardiac states with wearable photoplethysmograms and implications for out-of-hospital cardiac arrest detection
Scientific Reports,
Journal Year:
2024,
Volume and Issue:
14(1)
Published: Oct. 5, 2024
Out-of-hospital
cardiac
arrest
(OHCA)
is
a
global
health
problem
affecting
approximately
4.4
million
individuals
yearly.
OHCA
has
poor
survival
rate,
specifically
when
unwitnessed
(accounting
for
up
to
75%
of
cases).
Rapid
recognition
can
significantly
improve
survival,
and
consumer
wearables
with
continuous
cardiopulmonary
monitoring
capabilities
hold
potential
"witness"
activate
emergency
services.
In
this
study,
we
used
an
arterial
occlusion
model
simulate
investigated
the
ability
infrared
photoplethysmogram
(PPG)
sensors,
often
utilized
in
wearable
devices,
differentiate
normal
pulsation,
pulseless
(i.e.,
resembling
arrest),
non-physiologic
off-body)
states.
Across
classification
models
trained
evaluated
on
three
anatomical
locations,
higher
performances
were
observed
finger
(macro
average
F1-score
0.964
fingertip
0.954
base)
compared
wrist
0.837).
The
wrist-based
model,
which
was
using
all
PPG
measurements,
including
both
high-
low-quality
recordings,
achieved
macro
precision
recall
0.922
0.800,
respectively.
This
represents
most
common
form
factor
wearables,
could
only
capture
about
43.8%
events.
However,
tested
exclusively
high-quality
recordings
outcomes
0.975
fingertip,
0.973
base,
0.934
wrist).
had
highest
performance
pulselessness
from
pulsation
off-body
measurements
0.978
0.972,
able
identify
93.7%
states
event),
0.4%
false
positive
rate.
All
relied
combination
time-,
power
spectral
density
(PSD)-,
frequency-domain
features
cardiac,
recordings.
our
best
represented
idealized
detection
condition,
relying
ensuring
data
training
evaluation
machine
learning
algorithms.
While
90.7%
considered
high
quality,
53.2%
passed
quality
criteria.
Our
findings
have
implications
adapting
provide
detection,
involving
advancements
hardware
software
ensure
real-world
settings,
as
well
development
factors
that
enable
acquisition
more
consistently.
Given
these
improvements,
demonstrate
feasibly
be
made
available
anyone
PPG-based
wearables.
Language: Английский
The association of non-prescription drug use preceding out-of-hospital cardiac arrest and clinical outcomes
Resuscitation,
Journal Year:
2024,
Volume and Issue:
202, P. 110313 - 110313
Published: July 10, 2024
Language: Английский
Wearable devices for out‐of‐hospital cardiac arrest: A population survey on the willingness to adhere
Journal of the American College of Emergency Physicians Open,
Journal Year:
2024,
Volume and Issue:
5(5)
Published: Aug. 25, 2024
When
an
out-of-hospital
cardiac
arrest
(OHCA)
occurs,
the
first
step
in
chain
of
survival
is
detection.
However,
75%
OHCAs
are
unwitnessed,
representing
largest
barrier
to
activating
survival.
Wearable
devices
have
potential
be
"artificial
bystanders,"
detecting
OHCA
and
alerting
9-1-1.
We
sought
understand
factors
impacting
users'
willingness
for
continuous
use
a
wearable
device
through
online
survey
inform
future
these
systems
automated
Language: Английский