Life,
Год журнала:
2024,
Номер
14(11), С. 1488 - 1488
Опубликована: Ноя. 15, 2024
The
purpose
of
this
research
is
to
contribute
the
development
approaches
for
classification
and
segmentation
various
gastrointestinal
(GI)
cancer
diseases,
such
as
dyed
lifted
polyps,
resection
margins,
esophagitis,
normal
cecum,
pylorus,
Z
line,
ulcerative
colitis.
This
relevant
essential
because
current
challenges
related
absence
efficient
diagnostic
tools
early
diagnostics
GI
cancers,
which
are
fundamental
improving
diagnosis
these
common
diseases.
To
address
above
challenges,
we
propose
a
new
hybrid
model,
U-MaskNet,
combination
U-Net
Mask
R-CNN
models.
Here,
utilized
pixel-wise
instance
segmentation,
together
forming
solution
classifying
segmenting
cancer.
Kvasir
dataset,
includes
8000
endoscopic
images
validate
proposed
methodology.
experimental
results
clearly
demonstrated
that
novel
model
provided
superior
compared
other
well-known
models,
DeepLabv3+,
FCN,
DeepMask,
well
improved
performance
state-of-the-art
(SOTA)
including
LeNet-5,
AlexNet,
VGG-16,
ResNet-50,
Inception
Network.
quantitative
analysis
revealed
our
outperformed
achieving
precision
98.85%,
recall
98.49%,
F1
score
98.68%.
Additionally,
achieved
Dice
coefficient
94.35%
IoU
89.31%.
Consequently,
developed
increased
accuracy
reliability
in
detecting
cancer,
it
was
proven
can
potentially
be
used
process
and,
consequently,
patient
care
clinical
environment.
work
highlights
benefits
integrating
opening
way
further
medical
image
segmentation.
Applied Sciences,
Год журнала:
2025,
Номер
15(3), С. 1132 - 1132
Опубликована: Янв. 23, 2025
Reliably
detecting
COVID-19
is
critical
for
diagnosis
and
disease
control.
However,
imbalanced
data
in
medical
datasets
pose
significant
challenges
machine
learning
models,
leading
to
bias
poor
generalization.
The
dataset
obtained
from
the
EPIVIGILA
system
Chilean
Epidemiological
Surveillance
Process
contains
information
on
over
6,000,000
patients,
but,
like
many
current
datasets,
it
suffers
class
imbalance.
To
address
this
issue,
we
applied
various
algorithms,
both
with
without
sampling
methods,
compared
them
using
different
classification
diagnostic
metrics
such
as
precision,
sensitivity,
specificity,
likelihood
ratio
positive,
odds
ratio.
Our
results
showed
that
applying
methods
improved
metric
values
contributed
models
better
Effectively
managing
crucial
reliable
diagnosis.
This
study
enhances
understanding
of
how
techniques
can
improve
reliability
contribute
patient
outcomes.
Diagnostics,
Год журнала:
2024,
Номер
14(4), С. 454 - 454
Опубликована: Фев. 19, 2024
In
recent
years,
there
has
been
growing
interest
in
the
use
of
computer-assisted
technology
for
early
detection
skin
cancer
through
analysis
dermatoscopic
images.
However,
accuracy
illustrated
behind
state-of-the-art
approaches
depends
on
several
factors,
such
as
quality
images
and
interpretation
results
by
medical
experts.
This
systematic
review
aims
to
critically
assess
efficacy
challenges
this
research
field
order
explain
usability
limitations
highlight
potential
future
lines
work
scientific
clinical
community.
study,
was
carried
out
over
45
contemporary
studies
extracted
from
databases
Web
Science
Scopus.
Several
computer
vision
techniques
related
image
video
processing
diagnosis
were
identified.
context,
focus
process
included
algorithms
employed,
result
accuracy,
validation
metrics.
Thus,
yielded
significant
advancements
using
deep
learning
machine
algorithms.
Lastly,
establishes
a
foundation
research,
highlighting
contributions
opportunities
improve
effectiveness
learning.
Heliyon,
Год журнала:
2025,
Номер
11(2), С. e42119 - e42119
Опубликована: Янв. 1, 2025
Motion
disorders
affect
a
significant
portion
of
the
global
population.
While
some
symptoms
can
be
managed
with
medications,
these
treatments
often
impact
all
muscles
uniformly,
not
just
affected
ones,
leading
to
potential
side
effects
including
involuntary
movements,
confusion,
and
decreased
short-term
memory.
Currently,
there
is
no
dedicated
application
for
differentiating
healthy
from
abnormal
ones.
Existing
analysis
applications,
designed
other
purposes,
lack
essential
software
engineering
features
such
as
user-friendly
interface,
infrastructure
independence,
usability
learning
ability,
cloud
computing
capabilities,
AI-based
assistance.
This
research
proposes
computer-based
methodology
analyze
human
motion
differentiate
between
unhealthy
muscles.
First,
an
IoT-based
approach
proposed
digitize
using
smartphones
instead
hardly
accessible
wearable
sensors
markers.
The
data
then
simulated
neuromusculoskeletal
system.
An
agent-driven
modeling
method
ensures
naturalness,
accuracy,
interpretability
simulation,
incorporating
neuromuscular
details
Henneman's
size
principle,
action
potentials,
motor
units,
biomechanical
principles.
results
are
provided
medical
clinical
experts
aid
in
further
investigation.
Additionally,
deep
learning-based
ensemble
framework
assist
simulation
results,
offering
both
accuracy
interpretability.
A
graphical
interface
enhances
application's
usability.
Being
fully
cloud-based,
infrastructure-independent
accessed
on
smartphones,
PCs,
devices
without
installation.
strategy
only
addresses
current
challenges
treating
but
also
paves
way
simulations
by
considering
scientific
computational
requirements.
Life,
Год журнала:
2024,
Номер
14(7), С. 850 - 850
Опубликована: Июль 5, 2024
Background:
Surgical
site
infections
(SSIs)
represent
a
noteworthy
contributor
to
both
morbidity
and
mortality
in
the
context
of
patients
who
undergo
colorectal
surgery.
Several
risk
factors
have
been
identified;
however,
their
relative
significance
remains
uncertain.
Methods:
We
conducted
meta-analysis
observational
studies
from
inception
up
until
2023
that
investigated
for
SSIs
A
random-effects
model
was
used
pool
data
calculate
odds
ratio
(OR)
95%
confidence
interval
(CI)
each
factor.
Results:
Our
analysis
included
26
with
total
61,426
patients.
The
pooled
results
showed
male
sex
(OR
=
1.45),
body
mass
index
(BMI)
≥
25
kg/m2
1.09),
American
Society
Anesthesiologists
(ASA)
score
3
1.69),
were
all
independent
Conversely,
laparoscopic
surgery
0.70)
found
be
protective
Conclusions:
revealed
various
factors,
modifiable
non-modifiable,
associated
surgical
These
findings
emphasize
targeted
interventions,
including
optimizing
glycemic
control,
minimizing
blood
loss,
using
techniques
whenever
feasible
order
decrease
occurrence
this
particular
group
Diagnostics,
Год журнала:
2024,
Номер
14(16), С. 1746 - 1746
Опубликована: Авг. 12, 2024
Gastric
cancer
has
become
a
serious
worldwide
health
concern,
emphasizing
the
crucial
importance
of
early
diagnosis
measures
to
improve
patient
outcomes.
While
traditional
histological
image
analysis
is
regarded
as
clinical
gold
standard,
it
labour
intensive
and
manual.
In
recognition
this
problem,
there
been
rise
in
interest
use
computer-aided
diagnostic
tools
help
pathologists
with
their
efforts.
particular,
deep
learning
(DL)
emerged
promising
solution
sector.
However,
current
DL
models
are
still
restricted
ability
extract
extensive
visual
characteristics
for
correct
categorization.
To
address
limitation,
study
proposes
ensemble
models,
which
incorporate
capabilities
several
deep-learning
architectures
aggregate
knowledge
many
classification
performance,
allowing
more
accurate
efficient
gastric
detection.
determine
how
well
these
proposed
performed,
compared
them
other
works,
all
were
based
on
Histopathology
Sub-Size
Images
Database,
publicly
available
dataset
cancer.
This
research
demonstrates
that
achieved
high
detection
accuracy
across
sub-databases,
an
average
exceeding
99%.
Specifically,
ResNet50,
VGGNet,
ResNet34
performed
better
than
EfficientNet
VitNet.
For
80
×
80-pixel
sub-database,
exhibited
approximately
93%,
VGGNet
94%,
model
excelled
120
120-pixel
showed
99%
accuracy,
97%,
ResNet50
97%.
160
160-pixel
again
98%,
92%,
highlighting
model’s
superior
performance
resolutions.
Overall,
consistently
provided
three
sub-pixel
categories.
These
findings
show
may
successfully
detect
critical
from
smaller
patches
achieve
performance.
The
will
diagnose
using
histopathological
images,
leading
earlier
identification
higher
survival
rates.
Mathematics,
Год журнала:
2024,
Номер
12(18), С. 2808 - 2808
Опубликована: Сен. 11, 2024
Breast
cancer
is
one
of
the
most
lethal
and
widespread
diseases
affecting
women
worldwide.
As
a
result,
it
necessary
to
diagnose
breast
accurately
efficiently
utilizing
cost-effective
widely
used
methods.
In
this
research,
we
demonstrated
that
synthetically
created
high-quality
ultrasound
data
outperformed
conventional
augmentation
strategies
for
diagnosing
using
deep
learning.
We
trained
deep-learning
model
EfficientNet-B7
architecture
large
dataset
3186
images
acquired
from
multiple
publicly
available
sources,
as
well
10,000
generated
generative
adversarial
networks
(StyleGAN3).
The
was
five-fold
cross-validation
techniques
validated
four
metrics:
accuracy,
recall,
precision,
F1
score
measure.
results
showed
integrating
produced
into
training
set
increased
classification
accuracy
88.72%
92.01%
based
on
score,
demonstrating
power
models
expand
improve
quality
datasets
in
medical-imaging
applications.
This
larger
comprising
synthetic
significantly
improved
its
performance
by
more
than
3%
over
genuine
with
common
augmentation.
Various
procedures
were
also
investigated
set’s
diversity
representativeness.
research
emphasizes
relevance
modern
artificial
intelligence
machine-learning
technologies
medical
imaging
providing
an
effective
strategy
categorizing
images,
which
may
lead
diagnostic
optimal
treatment
options.
proposed
are
highly
promising
have
strong
potential
future
clinical
application
diagnosis
cancer.