International Journal of Imaging Systems and Technology,
Journal Year:
2024,
Volume and Issue:
35(1)
Published: Dec. 14, 2024
ABSTRACT
To
alleviate
the
burden
of
manual
annotation,
there
are
numerous
excellent
segmentation
models
for
images
being
developed.
However,
performance
these
data‐driven
is
frequently
constrained
by
availability
samples
sizes
pair
medical
and
annotations.
Therefore,
to
address
this
challenge,
study
introduces
image
augmentation
diffusion
model
(MEDSAD).
MEDSAD
solves
problem
annotation
scarcity
utilizing
a
given
simple
generate
paired
images.
improve
stability,
we
used
traditional
study.
exert
better
control
over
texture
synthesis
in
generated
MEDSAD,
style
injection
(TSI)
mechanism
introduced.
Additionally,
propose
feature
frequency
domain
attention
(FFDA)
module
mitigate
adverse
effects
high‐frequency
noise
during
generation.
The
efficacy
substantiated
through
validation
three
distinct
tasks
encompassing
magnetic
resonance
(MR)
ultrasound
(US)
imaging
modalities,
focusing
on
breast
tumors,
brain
nerve
structures.
findings
demonstrate
model's
proficiency
synthesizing
pairs
based
provided
annotations,
thereby
facilitating
notable
subsequent
tasks.
Moreover,
improvement
becomes
greater
as
quantity
synthetic
available
data
increases.
This
underscores
robust
generalization
capability
intrinsic
model,
potentially
offering
avenues
future
explorations
training
research.
Photodiagnosis and Photodynamic Therapy,
Journal Year:
2024,
Volume and Issue:
46, P. 104046 - 104046
Published: March 11, 2024
:
This
study
explores
the
intricate
connections
between
choroidal
vascular
index
(CVI)
and
non-invasive
ultrasonographic
atherosclerosis
predictors,
shedding
light
on
potential
links
ocular
dynamics
systemic
cardiovascular
health.
We
conducted
a
cross-sectional
analysis
of
81
participants,
assessing
CVI,
intima-media
thickness
(IMT),
extra-media
(EMT),
PATIMA
index.
The
presence
coronary
artery
disease
(CAD)
was
also
evaluated.
Statistical
methods
included
descriptive
statistics,
t-tests
for
group
comparisons,
Spearman
correlation
analysis,
receiver
operating
characteristic
(ROC)
curve
analysis.
Our
findings
revealed
that
patients
with
CAD
had
lower
CVI
values
compared
to
those
without
CAD,
underscoring
association
CAD.
Significant
negative
correlations
were
observed
IMT,
EMT,
PATIMA,
ROC
identified
optimal
cutoff
hypertension
detection,
showcasing
its
as
diagnostic
marker.
results
align
existing
literature
changes,
supporting
notion
may
be
promising
indicator
conditions.
contributes
broader
understanding
relationships
health,
providing
foundation
future
research
clinical
applications.
suggests
holds
relevance
marker
identifying
conditions,
offering
insights
into
fields
neurology,
physical
therapy,
rehabilitation.
Addressing
limitations,
this
encourages
further
investigation
multifaceted
predictors.
Scientific Reports,
Journal Year:
2024,
Volume and Issue:
14(1)
Published: April 27, 2024
Abstract
Medical
image
segmentation
is
a
key
task
in
computer
aided
diagnosis.
In
recent
years,
convolutional
neural
network
(CNN)
has
made
some
achievements
medical
segmentation.
However,
the
convolution
operation
can
only
extract
features
fixed
size
region
at
time,
which
leads
to
loss
of
features.
The
recently
popular
Transformer
global
modeling
capabilities,
but
it
does
not
pay
enough
attention
local
information
and
cannot
accurately
segment
edge
details
target
area.
Given
these
issues,
we
proposed
dynamic
regional
(DRA-Net).
Different
from
above
methods,
first
measures
similarity
concentrates
on
different
regions.
this
way,
adaptively
select
scopes
for
feature
extraction,
reducing
loss.
Then,
interaction
carried
out
better
learn
details.
At
same
also
design
ordered
shift
multilayer
perceptron
(MLP)
blocks
enhance
communication
within
regions,
further
enhancing
network’s
ability
After
several
experiments,
results
indicate
that
our
produces
more
accurate
performance
compared
other
CNN
based
networks.
Bioengineering,
Journal Year:
2023,
Volume and Issue:
10(6), P. 712 - 712
Published: June 12, 2023
In
recent
years,
UNet
and
its
improved
variants
have
become
the
main
methods
for
medical
image
segmentation.
Although
these
models
achieved
excellent
results
in
segmentation
accuracy,
their
large
number
of
network
parameters
high
computational
complexity
make
it
difficult
to
achieve
real-time
therapy
diagnosis
rapidly.
To
address
this
problem,
we
introduce
a
lightweight
(LcmUNet)
based
on
CNN
MLP.
We
designed
LcmUNet's
structure
terms
model
performance,
parameters,
complexity.
The
first
three
layers
are
convolutional
layers,
last
two
MLP
layers.
convolution
part,
propose
an
LDA
module
that
combines
asymmetric
convolution,
depth-wise
separable
attention
mechanism
reduce
while
maintaining
strong
feature-extraction
capability.
LMLP
helps
enhance
contextual
information
focusing
local
improves
accuracy
inference
speed.
This
also
covers
skip
connections
between
encoder
decoder
at
various
levels.
Our
achieves
accurately
extensive
experiments.
With
only
1.49
million
without
pre-training,
LcmUNet
demonstrated
impressive
performance
different
datasets.
On
ISIC2018
dataset,
IoU
85.19%,
92.07%
recall,
92.99%
precision.
BUSI
63.99%,
79.96%
76.69%
Lastly,
Kvasir-SEG
81.89%,
88.93%
91.79%
