The Role of Artificial Intelligence in Epiretinal Membrane Care: A Scoping Review
Ophthalmology Science,
Год журнала:
2024,
Номер
unknown, С. 100689 - 100689
Опубликована: Дек. 1, 2024
Язык: Английский
Application of Multimodal Imaging in the Diagnosis and Treatment of Epiretinal Membrane
Advances in Clinical Medicine,
Год журнала:
2025,
Номер
15(01), С. 676 - 683
Опубликована: Янв. 1, 2025
Язык: Английский
Performance of artificial intelligence-based models for epiretinal membrane diagnosis: A systematic review and meta-analysis
American Journal of Ophthalmology,
Год журнала:
2025,
Номер
unknown
Опубликована: Май 1, 2025
Epiretinal
membrane
(ERM)
can
impair
central
vision
by
forming
a
pre-retinal
fibrous
layer
on
the
inner
retina.
Artificial
intelligence
(AI)-based
tools
may
streamline
ERM
diagnosis,
but
their
overall
performance
and
factors
affecting
accuracy
require
evaluation.
With
an
aging
population,
prevalence
is
expected
to
rise,
placing
increased
demands
clinical
resources.
Early
detection
via
AI
models
could
expedite
reduce
subjective
errors,
guide
timely
surgical
intervention.
This
systematic
review
meta-analysis
evaluates
pooled
diagnostic
of
for
detecting
identifies
study-
model-level
influencing
performance.
Systematic
Review
Meta-Analysis
METHODS:
Comprehensive
searches
were
conducted
in
Medline,
Embase,
Cochrane
Library,
Web
Science,
preprint
databases
from
inception
June
2024.
Included
studies
evaluated
diagnosis.
Study
quality
risk
bias
assessed
using
Quality
Assessment
Diagnostic
Accuracy
Studies
2
(QUADAS-2)
tool.
A
random-effects
model
was
applied
pool
accuracy,
sensitivity,
specificity,
odds
ratio.
Subgroup
analyses
explored
The
study
protocol
registered
with
International
Prospective
Register
Reviews
(PROSPERO
-
CRD42024563571).
Of
379
articles
screened,
26
met
inclusion
criteria,
19
contributed
meta-analysis.
settings
predominantly
hospital-based
(76.9%),
some
academic
computer
biomedical
science
departments
(15.4%)
community
centers
(7.7%).
assessments
suggested
low
or
unclear
applicability
concerns
95%
studies.
sensitivity
90.1%
(95%
CI:
85.8-93.2),
specificity
95.7%
88.8-95.2).
analysis
showed
higher
(97.1%,
96.0-97.9)
color
fundus
photographs
than
optical
coherence
tomography
scans,
which
had
92.6%
External
validation
performed
26.9%
All
included
used
expert
human
grading
as
reference
standard,
25
(96.2%)
based
same
imaging
modality
input.
proportion
cases
development
datasets
varied
across
studies,
particularly
between
single-disease
multiclass
models.
demonstrate
high
ERM.
However,
limited
external
variability
methodologies
limits
direct
comparison
real-world
applicability.
Future
work
should
standardize
reporting
practices,
improve
data
interoperability,
develop
prediction
track
disease
progression
determine
optimal
timing.
Язык: Английский
Haemorrhage diagnosis in colour fundus images using a fast-convolutional neural network based on a modified U-Net
R. Sathiyaseelan,
R. Krishnamoorthy,
Ramesh Ramamoorthy
и другие.
Network Computation in Neural Systems,
Год журнала:
2024,
Номер
unknown, С. 1 - 22
Опубликована: Фев. 12, 2024
Retinal
haemorrhage
stands
as
an
early
indicator
of
diabetic
retinopathy,
necessitating
accurate
detection
for
timely
diagnosis.
Addressing
this
need,
study
proposes
enhanced
machine-based
diagnostic
test
retinopathy
through
updated
UNet
framework,
adept
at
scrutinizing
fundus
images
signs
retinal
haemorrhages.
The
customized
underwent
GPU
training
using
the
IDRiD
database,
validated
against
publicly
available
DIARETDB1
and
datasets.
Emphasizing
complexity
segmentation,
employed
preprocessing
techniques,
augmenting
image
quality
data
integrity.
Subsequently,
trained
neural
network
showcased
a
remarkable
performance
boost,
accurately
identifying
regions
with
80%
sensitivity,
99.6%
specificity,
98.6%
accuracy.
experimental
findings
solidify
network's
reliability,
showcasing
potential
to
alleviate
ophthalmologists'
workload
significantly.
Notably,
achieving
Intersection
over
Union
(IoU)
76.61%
Dice
coefficient
86.51%
underscores
system's
competence.
study's
outcomes
signify
substantial
enhancements
in
diagnosing
critical
conditions,
promising
profound
improvements
accuracy
efficiency,
thereby
marking
significant
advancement
automated
retinopathy.
Язык: Английский
A Novel Hybrid Retinal Blood Vessel Segmentation Algorithm for Enlarging the Measuring Range of Dual-Wavelength Retinal Oximetry
Photonics,
Год журнала:
2023,
Номер
10(7), С. 722 - 722
Опубликована: Июнь 24, 2023
The
non-invasive
measurement
of
hemoglobin
oxygen
saturation
(SO2)
in
retinal
vessels
is
based
on
spectrophotometry
and
the
absorption
spectral
characteristics
tissue.
dual-wavelength
images
are
simultaneously
captured
via
oximetry.
SO2
calculated
by
processing
a
series
calculating
optic
density
ratio
two
images.
However,
existing
research
focused
thick
high-clarity
region
thin
low-clarity
could
provide
significant
information
for
detection
diagnosis
neovascular
diseases.
To
this
end,
we
proposed
novel
hybrid
vessel
segmentation
algorithm.
Firstly,
median
filter
was
employed
image
denoising.
Secondly,
high-
carried
out
clarity
histogram.
areas
were
segmented
after
implementing
Gaussian
filter,
matched
morphological
segmentation.
Additionally,
using
guided
filtering,
dynamic
threshold
Finally,
results
obtained
through
merger
operations.
experimental
analysis
show
that
method
can
effectively
segment
extend
measuring
range
Язык: Английский