Journal of Sleep Research,
Journal Year:
2024,
Volume and Issue:
unknown
Published: July 18, 2024
Summary
Developing
a
convenient
detection
method
is
important
for
diagnosing
and
treating
obstructive
sleep
apnea.
Considering
availability
medical
reliability,
we
established
deep‐learning
model
that
uses
single‐lead
electrocardiogram
signals
apnea
severity
assessment.
The
consisted
of
signal
preprocessing,
feature
extraction,
time–frequency
domain
information
fusion,
classification
segments.
A
total
375
patients
who
underwent
polysomnography
were
included.
obtained
by
used
to
train,
validate
test
the
model.
Moreover,
proposed
performance
on
public
dataset
was
compared
with
findings
previous
studies.
In
set,
accuracy
per‐segment
per‐recording
82.55%
85.33%,
respectively.
values
mild,
moderate
severe
69.33%,
74.67%
dataset,
91.66%.
Bland–Altman
plot
revealed
consistency
true
apnea–hypopnea
index
predicted
index.
We
confirmed
feasibility
evaluation
in
both
hospital
datasets.
high
apnea,
especially
those
BMC Medical Informatics and Decision Making,
Journal Year:
2023,
Volume and Issue:
23(1)
Published: Oct. 19, 2023
Obstructive
sleep
apnea
(OSA)
is
a
globally
prevalent
disease
with
complex
diagnostic
method.
Severe
OSA
associated
multi-system
dysfunction.
We
aimed
to
develop
an
interpretable
machine
learning
(ML)
model
for
predicting
the
risk
of
severe
and
analyzing
factors
based
on
clinical
characteristics
questionnaires.This
was
retrospective
study
comprising
1656
subjects
who
presented
underwent
polysomnography
(PSG)
between
2018
2021.
A
total
23
variables
were
included,
after
univariate
analysis,
15
selected
further
preprocessing.
Six
types
classification
models
used
evaluate
ability
predict
OSA,
namely
logistic
regression
(LR),
gradient
boosting
(GBM),
extreme
(XGBoost),
adaptive
(AdaBoost),
bootstrapped
aggregating
(Bagging),
multilayer
perceptron
(MLP).
All
area
under
receiver
operating
characteristic
curve
(AUC)
calculated
as
performance
metric.
also
drew
SHapley
Additive
exPlanations
(SHAP)
plots
interpret
predictive
results
analyze
relative
importance
factors.
An
online
calculator
developed
estimate
in
individuals.Among
enrolled
subjects,
61.47%
(1018/1656)
diagnosed
OSA.
Multivariate
LR
analysis
showed
that
10
independent
The
GBM
best
(AUC
=
0.857,
accuracy
0.766,
sensitivity
0.798,
specificity
0.734).
model.
Finally,
waist
circumference,
neck
Epworth
Sleepiness
Scale,
age,
Berlin
questionnaire
revealed
by
SHAP
plot
top
five
critical
contributing
diagnosis
Additionally,
two
typical
cases
analyzed
contribution
each
variable
outcome
prediction
single
patient.We
established
six
using
ML
algorithms.
Among
them,
performed
best.
facilitates
individualized
assessment
strategies
patients
suspected
This
will
help
identify
early
possible
ensure
their
timely
treatment.Retrospectively
registered.
Journal of Personalized Medicine,
Journal Year:
2024,
Volume and Issue:
14(6), P. 598 - 598
Published: June 4, 2024
Personalized
sleep
medicine
represents
a
transformative
shift
in
healthcare,
emphasizing
individualized
approaches
to
optimizing
health,
considering
the
bidirectional
relationship
between
and
health.
This
field
moves
beyond
conventional
methods,
tailoring
care
unique
physiological
psychological
needs
of
individuals
improve
quality
manage
disorders.
Key
this
approach
is
consideration
diverse
factors
like
genetic
predispositions,
lifestyle
habits,
environmental
factors,
underlying
health
conditions.
enables
more
accurate
diagnoses,
targeted
treatments,
proactive
management.
Technological
advancements
play
pivotal
role
field:
wearable
devices,
mobile
applications,
advanced
diagnostic
tools
collect
detailed
data
for
continuous
monitoring
analysis.
The
integration
machine
learning
artificial
intelligence
enhances
interpretation,
offering
personalized
treatment
plans
based
on
individual
profiles.
Moreover,
research
circadian
rhythms
physiology
advancing
our
understanding
sleep’s
impact
overall
next
generation
technology
will
integrate
seamlessly
with
IoT
smart
home
systems,
facilitating
holistic
environment
Telemedicine
virtual
healthcare
platforms
increase
accessibility
specialized
care,
especially
remote
areas.
Advancements
also
focus
integrating
various
sources
comprehensive
assessments
treatments.
Genomic
molecular
could
lead
breakthroughs
disorders,
informing
highly
plans.
Sophisticated
methods
stage
estimation,
including
techniques,
are
improving
precision.
Computational
models,
particularly
conditions
obstructive
apnea,
enabling
patient-specific
strategies.
future
likely
involve
cross-disciplinary
collaborations,
cognitive
behavioral
therapy
mental
interventions.
Public
awareness
education
about
approaches,
alongside
updated
regulatory
frameworks
security
privacy,
essential.
Longitudinal
studies
provide
insights
into
evolving
patterns,
further
refining
approaches.
In
conclusion,
revolutionizing
disorder
treatment,
leveraging
characteristics
technologies
improved
diagnosis,
towards
marks
significant
advancement
enhancing
life
those
Immunity Inflammation and Disease,
Journal Year:
2024,
Volume and Issue:
12(10)
Published: Oct. 1, 2024
Abstract
Background
Over
the
past
decades,
it
has
become
increasingly
evident
that
sleep
disturbance
contributes
to
inflammation‐mediated
disease,
including
depression,
mainly
through
activation
of
innate
immune
system
and
an
increased
risk
infections.
Methods
A
comprehensive
literature
search
was
performed
in
PubMed
identify
relevant
research
findings
field
immunity,
inflammation
infections,
with
a
focus
on
translational
from
5
years.
Results
Physiological
is
characterized
by
dynamic
interplay
between
architecture,
marked
immunity
T
helper
1
(Th1)
‐mediated
early
phase,
transitioning
2
(Th2)
response
dominating
late
sleep.
Chronic
disturbances
are
associated
enhanced
elevated
while
other
inflammatory
diseases
may
also
be
affected.
Conversely,
infection
can
disrupt
patterns
architecture.
This
narrative
review
summarizes
current
data
complex
relationships
sleep,
highlighting
aspects.
The
bidirectional
nature
these
interactions
addressed
within
specific
conditions
such
as
apnea,
HIV,
Furthermore,
technical
developments
potential
accelerate
our
understanding
identified,
advances
wearable
devices,
artificial
intelligence,
omics
technology.
By
integrating
tools,
novel
biomarkers
therapeutic
targets
for
sleep‐related
dysregulation
identified.
Conclusion
underscores
importance
addressing
imbalance
related
improve
disease
outcomes.
Sleep Medicine Reviews,
Journal Year:
2022,
Volume and Issue:
63, P. 101611 - 101611
Published: Feb. 17, 2022
Sleep
is
characterized
by
an
intricate
variation
of
brain
activity
over
time.
Measuring
these
temporal
sleep
dynamics
relevant
for
elucidating
healthy
and
pathological
mechanisms.
The
rapidly
increasing
possibilities
obtaining
processing
registrations
have
led
to
abundance
data,
which
can
be
challenging
analyze
interpret.
This
review
provides
a
structured
overview
approaches
represent
dynamics,
categorized
based
on
the
way
source
data
compressed.
For
each
category
representations,
we
describe
advantages
disadvantages.
Standard
human-defined
30-s
stages
standardization
interpretability.
Alternative
representations
are
less
standardized
but
offer
higher
resolution
(in
case
microstructural
events
such
as
spindles),
or
reflect
non-categorical
information
(for
example
spectral
power
analysis).
Machine-learned
additional
possibilities:
automated
useful
handling
large
quantities
while
alternative
obtained
from
clustering
data-driven
features
could
aid
finding
new
patterns
possible
clinical
interpretations.
While
newly
developed
may
insights,
they
difficult
interpret
in
context.
Therefore,
there
should
always
balance
between
developing
sophisticated
analysis
techniques
maintaining
explainability.
Sensors,
Journal Year:
2023,
Volume and Issue:
23(8), P. 3973 - 3973
Published: April 13, 2023
Sleep
disorders
can
impact
daily
life,
affecting
physical,
emotional,
and
cognitive
well-being.
Due
to
the
time-consuming,
highly
obtrusive,
expensive
nature
of
using
standard
approaches
such
as
polysomnography,
it
is
great
interest
develop
a
noninvasive
unobtrusive
in-home
sleep
monitoring
system
that
reliably
accurately
measure
cardiorespiratory
parameters
while
causing
minimal
discomfort
user’s
sleep.
We
developed
low-cost
Out
Center
Testing
(OCST)
with
low
complexity
parameters.
tested
validated
two
force-sensitive
resistor
strip
sensors
under
bed
mattress
covering
thoracic
abdominal
regions.
Twenty
subjects
were
recruited,
including
12
males
8
females.
The
ballistocardiogram
signal
was
processed
4th
smooth
level
discrete
wavelet
transform
2nd
order
Butterworth
bandpass
filter
heart
rate
respiration
rate,
respectively.
reached
total
error
(concerning
reference
sensors)
3.24
beats
per
minute
2.32
rates
for
For
females,
errors
3.47
2.68,
2.33,
verified
reliability
applicability
system.
It
showed
minor
dependency
on
sleeping
positions,
one
major
cumbersome
measurements.
identified
sensor
region
optimal
configuration
measurement.
Although
testing
healthy
regular
patterns
promising
results,
further
investigation
required
bandwidth
frequency
validation
larger
groups
subjects,
patients.
Nature and Science of Sleep,
Journal Year:
2025,
Volume and Issue:
Volume 17, P. 69 - 79
Published: Jan. 1, 2025
The
incidence
of
insomnia
in
cancer
patients
is
significantly
higher
than
the
general
population.
Chronic
imposes
pronounced
physical
and
psychological
burdens
on
patients,
affecting
their
quality
life
survival
rate.
This
study
aims
to
investigate
further
analyze
potentially
related
factors.
Oncology
outpatients
treated
at
Fudan
University
Shanghai
Cancer
Center
were
consecutively
recruited.
Demographic
information
clinical
features,
such
as
type
treatment
status,
collected.
Insomnia
was
assessed
using
Severity
Index
(ISI).
A
total
146
participated
study,
with
majority
suffering
from
breast
tumors
(40.4%),
gastrointestinal
tract
(18.5%),
endocrine
(5.8%).
Among
these
25
(17.1%)
did
not
report
insomnia,
69
(47.3%)
had
subclinical
52
(35.6%)
reached
level
insomnia.
Older
aged
41-50
years
(Estimate
=
-3.49,
95%
CI,
-6.99
0.00,
p
0.05)
those
education
levels
-2.72,
-4.88
-0.55,
0.01)
less
likely
have
ISI
scores.
In
contrast,
undergoing
chemotherapy
3.86,
0.53
7.19,
0.02)
associated
Gender,
age,
education,
modalities
correlated
subitem
prevalence
oncology
gender,
tumor
type,
modality.
Screening
interventions
for
should
be
emphasized
whole-course
management
patients.
