Journal of Radiation Research and Applied Sciences,
Год журнала:
2023,
Номер
16(3), С. 100602 - 100602
Опубликована: Июль 1, 2023
Gastric
cancer
is
a
kind
of
tumor
with
high
morbidity
and
mortality,
which
seriously
threatens
people's
health
life.
It
great
significance
to
study
the
early
diagnosis
screening
for
improving
cure
rate
cancer,
prolonging
survival
time
patients,
reducing
economic
mental
burden
patients.
Because
deep
convolutional
neural
networks
can
effectively
extract
features
images,
gooenet
AlexNet
models
perform
wonderful
image
classification,
they
are
selected
pathological
images
gastric
cancer.
Moreover,
GooleNet
model
optimized
make
it
more
targeted
at
medical
not
only
ensures
diagnostic
accuracy,
but
also
significantly
reduces
computational
burden.
The
improved
has
characteristics
two
kinds
network
structure
same
time,
sections,
sensitivity
section
recognition.
results
show
that
splendid
accuracy
up
97.
61%,
specificity
99.
47
percent.
diagnose
accurately,
reduce
possibility
misdiagnosis
missed
due
doctors'
personal
reasons,
help
nurses
care
monitor
making
whole
treatment
process
intelligent
safe.
Frontiers in Cell and Developmental Biology,
Год журнала:
2025,
Номер
13
Опубликована: Май 21, 2025
Generative
adversarial
networks
(GANs)
were
employed
to
predict
the
morphology
of
OBL
before
femtosecond
laser
scanning
during
SMILE.
A
retrospective
cross-sectional
analysis
was
conducted
on
4,442
eyes
from
2,265
patients
who
underwent
SMILE
surgery
at
Ophthalmic
Center
Second
Affiliated
Hospital
Nanchang
University
between
June
2021
and
August
2022.
Surgical
videos,
preoperative
panoramic
corneal
images,
intraoperative
images
collected.
The
dataset
randomly
split
into
a
training
set
3,998
test
444
for
model
development
evaluation,
respectively.
Structural
similarity
index
(SSIM)
peak
signal-to-noise
ratio
(PSNR)
used
quantitatively
assess
image
quality.
accuracy
predictions
also
compared
across
different
models.
Seven
GAN
models
developed.
Among
them,
incorporating
residual
structure
Transformer
module
within
Pix2pix
framework
exhibited
best
predictive
performance.
This
model's
prediction
demonstrated
high
consistency
with
actual
(SSIM
=
0.67,
PSNR
26.02).
Trans-Pix2Pix
0.66,
25.76),
Res-Pix2Pix
0.65,
23.08),
Pix2Pix
0.64,
22.97),
Pix2PixHD
0.63,
23.46),
DCGAN
0.58,
20.46)
slightly
lower,
while
CycleGAN
0.51,
18.30)
showed
least
favorable
results.
developed
predicting
based
demonstrates
effective
capabilities
offers
valuable
insights
ophthalmologists
in
surgical
planning.
Journal of Radiation Research and Applied Sciences,
Год журнала:
2023,
Номер
16(3), С. 100603 - 100603
Опубликована: Июнь 1, 2023
To
introduce
an
improved
method
of
digital
subtraction
angiography
image
segmentation
based
on
a
multi-scale
Hessian
matrix,
to
accurately
segment
vascular
images
before
and
after
coronary
stent
implantation,
which
will
have
strong
applications
in
medical
diagnostics
clinical
nursing
patients
who
underwent
stents
implantation.
Firstly,
vessel
edge
enhancement
algorithm
is
proposed,
enhances
the
gradient
edge,
not
only
makes
obtained
smoother,
but
also
visual
effect
intersection
vessels;
secondly,
introduces
noise
filtering
morphology,
can
detect
remove
linear
similar
blood
finally,
view
problem
that
each
DSA
sequence
show
part
vessels,
conducive
observing
overall
state
fusion
designed
display
vessels
same
image.
The
be
understood
as
whole
one
Compared
with
literature,
overlap
rate
increased
by
0.0089,
mis-segmentation
decreased
0.0334.
In
quantitative
analysis,
this
paper
improves
overlapping
reduces
rate.
For
comparison
effects,
segmented
higher
clarity,
helpful
for
doctors
nurses
comprehensively
synthesize
information.
There
more
objective
basis
preoperative
diagnosis
postoperative
rehabilitation
angiographic
Journal of Radiation Research and Applied Sciences,
Год журнала:
2023,
Номер
16(3), С. 100602 - 100602
Опубликована: Июль 1, 2023
Gastric
cancer
is
a
kind
of
tumor
with
high
morbidity
and
mortality,
which
seriously
threatens
people's
health
life.
It
great
significance
to
study
the
early
diagnosis
screening
for
improving
cure
rate
cancer,
prolonging
survival
time
patients,
reducing
economic
mental
burden
patients.
Because
deep
convolutional
neural
networks
can
effectively
extract
features
images,
gooenet
AlexNet
models
perform
wonderful
image
classification,
they
are
selected
pathological
images
gastric
cancer.
Moreover,
GooleNet
model
optimized
make
it
more
targeted
at
medical
not
only
ensures
diagnostic
accuracy,
but
also
significantly
reduces
computational
burden.
The
improved
has
characteristics
two
kinds
network
structure
same
time,
sections,
sensitivity
section
recognition.
results
show
that
splendid
accuracy
up
97.
61%,
specificity
99.
47
percent.
diagnose
accurately,
reduce
possibility
misdiagnosis
missed
due
doctors'
personal
reasons,
help
nurses
care
monitor
making
whole
treatment
process
intelligent
safe.
