Applied Sciences,
Год журнала:
2024,
Номер
14(17), С. 7447 - 7447
Опубликована: Авг. 23, 2024
The
recent
increase
in
the
prevalence
of
skin
cancer,
along
with
its
significant
impact
on
individuals’
lives,
has
garnered
attention
many
researchers
field
deep
learning
models,
especially
following
promising
results
observed
using
these
models
medical
field.
This
study
aimed
to
develop
a
system
that
can
accurately
diagnose
one
three
types
cancer:
basal
cell
carcinoma
(BCC),
melanoma
(MEL),
and
nevi
(NV).
Additionally,
it
emphasizes
importance
image
quality,
as
studies
focus
quantity
images
used
learning.
In
this
study,
transfer
was
employed
pre-trained
VGG-16
model
alongside
dataset
sourced
from
Kaggle.
Three
were
trained
while
maintaining
same
hyperparameters
script
ensure
fair
comparison.
However,
data
train
each
varied
observe
specific
effects
hypothesize
about
quality
within
highest
validation
score
selected
for
further
testing
separate
test
dataset,
which
had
not
seen
before,
evaluate
model’s
performance
accurately.
work
contributes
existing
body
research
by
demonstrating
critical
role
enhancing
diagnostic
accuracy,
providing
comprehensive
evaluation
cancer
detection
offering
insights
guide
future
improvements
World Journal of Advanced Research and Reviews,
Год журнала:
2024,
Номер
21(1), С. 161 - 171
Опубликована: Янв. 4, 2024
The
rapid
increase
in
human
activities
is
causing
significant
damage
to
our
planet's
ecosystems,
necessitating
innovative
solutions
preserve
biodiversity
and
counteract
ecological
threats.
Artificial
Intelligence
(AI)
has
emerged
as
a
transformative
force,
providing
unparalleled
capabilities
for
environmental
monitoring
conservation.
This
research
paper
explores
the
applications
of
AI
ecosystem
management,
including
wildlife
tracking,
habitat
assessment,
analysis,
natural
disaster
prediction.
AI's
role
conservation
includes
resource
conservation,
species
identification.
algorithms
analyze
camera
trap
footage,
drone
imagery,
GPS
data
identify
estimate
population
sizes,
leading
improved
anti-poaching
efforts
enhanced
protection
diverse
species.
Habitat
assessment
involve
AI-powered
image
which
aids
assessing
forest
health,
detecting
deforestation,
identifying
areas
need
restoration.
Biodiversity
analysis
identification
are
achieved
through
that
acoustic
recordings,
DNA
(eDNA),
footage.
These
innovations
different
species,
assess
levels,
even
discover
new
or
endangered
flood
prediction
systems
provide
early
warnings,
empowering
communities
with
better
preparedness
evacuation
efforts.
Challenges,
such
quality
availability,
algorithmic
bias,
infrastructure
limitations,
acknowledged
opportunities
growth
improvement.
In
policy
regulation,
advocates
clear
frameworks
prioritizing
privacy
security,
transparency,
equitable
access.
Responsible
development
ethical
use
emphasized
foundational
pillars,
ensuring
integration
into
aligns
principles
fairness,
societal
benefit.
Advances in healthcare information systems and administration book series,
Год журнала:
2024,
Номер
unknown, С. 220 - 238
Опубликована: Фев. 14, 2024
This
chapter
explores
particle
swarm
optimization
(PSO)
in
the
rapidly
evolving
landscape
of
biomedical
technologies.
The
study
begins
by
introducing
fundamental
principles
PSO,
emphasizing
its
advantages
addressing
complex
problems
common
applications.
authors
delve
into
innovative
uses
PSO
various
fields,
including
image
enhancement,
data
clustering,
and
drug
development,
highlighting
how
contributes
to
more
accurate
diagnoses,
efficient
treatment
plans,
streamlined
research
methodologies.
Significantly,
this
identifies
emerging
opportunities
where
can
be
further
leveraged,
particularly
personalized
medicine
predictive
health
analytics,
suggesting
a
roadmap
for
future
development.
By
combining
theoretical
insights
with
practical
examples,
work
aims
provide
comprehensive
overview
PSO's
role
advancing
technologies,
offering
valuable
perspectives
researchers,
practitioners,
policymakers
field.
Deleted Journal,
Год журнала:
2024,
Номер
37(3), С. 1038 - 1053
Опубликована: Фев. 13, 2024
Abstract
Breast
microcalcifications
are
observed
in
80%
of
mammograms,
and
a
notable
proportion
can
lead
to
invasive
tumors.
However,
diagnosing
is
highly
complicated
error-prone
process
due
their
diverse
sizes,
shapes,
subtle
variations.
In
this
study,
we
propose
radiomic
signature
that
effectively
differentiates
between
healthy
tissue,
benign
microcalcifications,
malignant
microcalcifications.
Radiomic
features
were
extracted
from
proprietary
dataset,
composed
380
136
benign,
242
ROIs.
Subsequently,
two
distinct
signatures
selected
differentiate
tissue
(detection
task)
(classification
task).
Machine
learning
models,
namely
Support
Vector
Machine,
Random
Forest,
XGBoost,
employed
as
classifiers.
The
shared
for
both
tasks
was
then
used
train
multi-class
model
capable
simultaneously
classifying
healthy,
A
significant
overlap
discovered
the
detection
classification
signatures.
performance
models
promising,
with
XGBoost
exhibiting
an
AUC-ROC
0.830,
0.856,
0.876
classification,
respectively.
intrinsic
interpretability
features,
use
Mean
Score
Decrease
method
introspection,
enabled
models’
clinical
validation.
fact,
most
important
GLCM
Contrast,
FO
Minimum
Entropy,
compared
found
other
studies
on
breast
cancer.
Journal of Materials Chemistry A,
Год журнала:
2024,
Номер
unknown
Опубликована: Янв. 1, 2024
This
review
examines
the
integration
of
artificial
intelligence
with
nanogenerators
to
develop
self-powered,
adaptive
systems
for
applications
in
robotics,
wearables,
and
environmental
monitoring.
Frontiers in Medicine,
Год журнала:
2025,
Номер
12
Опубликована: Март 19, 2025
Computed
tomography
pulmonary
angiography
(CTPA)
is
an
essential
diagnostic
tool
for
identifying
embolism
(PE).
The
integration
of
AI
has
significantly
advanced
CTPA-based
PE
detection,
enhancing
accuracy
and
efficiency.
This
review
investigates
the
growing
role
in
diagnosis
using
CTPA
imaging.
examines
capabilities
algorithms,
particularly
deep
learning
models,
analyzing
images
detection.
It
assesses
their
sensitivity
specificity
compared
to
human
radiologists.
systems,
large
datasets
complex
neural
networks,
demonstrate
remarkable
proficiency
subtle
signs
PE,
aiding
clinicians
timely
accurate
diagnosis.
In
addition,
AI-powered
analysis
shows
promise
risk
stratification,
prognosis
prediction,
treatment
optimization
patients.
Automated
image
interpretation
quantitative
facilitate
rapid
triage
suspected
cases,
enabling
prompt
intervention
reducing
delays.
Despite
these
advancements,
several
limitations
remain,
including
algorithm
bias,
interpretability
issues,
necessity
rigorous
validation,
which
hinder
widespread
adoption
clinical
practice.
Furthermore,
integrating
into
existing
healthcare
systems
requires
careful
consideration
regulatory,
ethical,
legal
implications.
conclusion,
AI-driven
detection
presents
unprecedented
opportunities
enhance
precision
However,
addressing
associated
critical
safe
effective
implementation
routine
Successful
utilization
revolutionizing
care
necessitates
close
collaboration
among
researchers,
medical
professionals,
regulatory
organizations.
Applied Sciences,
Год журнала:
2024,
Номер
14(7), С. 2970 - 2970
Опубликована: Март 31, 2024
Sample
size
is
a
key
factor
in
bioequivalence
and
clinical
trials.
An
appropriately
large
sample
necessary
to
gain
valuable
insights
into
designated
population.
However,
sizes
lead
increased
human
exposure,
costs,
longer
time
for
completion.
In
previous
study,
we
introduced
the
idea
of
using
variational
autoencoders
(VAEs),
type
artificial
neural
network,
synthetically
create
studies.
this
work,
further
elaborate
on
expand
it
field
(BE)
A
computational
methodology
was
developed,
combining
Monte
Carlo
simulations
2
×
crossover
BE
trials
with
deep
learning
algorithms,
specifically
VAEs.
Various
scenarios,
including
variability
levels,
actual
size,
VAE-generated
difference
performance
between
two
pharmaceutical
products
under
comparison,
were
explored.
All
showed
that
incorporating
AI
generative
algorithms
creating
virtual
populations
has
many
advantages,
as
less
data
can
be
used
achieve
similar,
even
better,
results.
Overall,
work
shows
how
application
like
VAEs,
clinical/bioequivalence
studies
modern
tool
significantly
reduce
trial
completion
time.
Cancers,
Год журнала:
2025,
Номер
17(1), С. 116 - 116
Опубликована: Янв. 2, 2025
The
ICIBM
2023
marked
the
11th
annual
conference
of
its
kind,
with
recently
becoming
official
International
Association
for
Intelligent
Biology
and
Medicine
(IAIBM),
showcasing
cutting-edge
advancements
at
intersection
computation
biomedical
research
[...]
Journal of Imaging,
Год журнала:
2025,
Номер
11(1), С. 26 - 26
Опубликована: Янв. 17, 2025
Object
detection
in
images
is
a
fundamental
component
of
many
safety-critical
systems,
such
as
autonomous
driving,
video
surveillance
and
robotics.
Adversarial
patch
attacks,
being
easily
implemented
the
real
world,
provide
effective
counteraction
to
object
by
state-of-the-art
neural-based
detectors.
It
poses
serious
danger
various
fields
activity.
Existing
defense
methods
against
attacks
are
insufficiently
effective,
which
underlines
need
develop
new
reliable
solutions.
In
this
manuscript,
we
propose
method
helps
increase
robustness
neural
network
systems
input
adversarial
images.
The
proposed
consists
Deep
Convolutional
Neural
Network
reconstruct
benign
image
from
one;
Calculating
Maximum
Error
block
highlight
mismatches
between
reconstructed
images;
Localizing
Anomalous
Fragments
extract
anomalous
regions
using
Isolation
Forest
algorithm
histograms
images'
fragments;
Clustering
Processing
group
evaluate
extracted
regions.
method,
based
on
anomaly
localization,
demonstrates
high
resistance
while
maintaining
quality
detection.
experimental
results
show
that
defending
attacks.
Using
YOLOv3
with
defensive
for
pedestrian
INRIAPerson
dataset
under
mAP50
metric
reaches
80.97%
compared
46.79%
without
method.
research
demonstrate
promising
improvement
security.
Advances in environmental engineering and green technologies book series,
Год журнала:
2025,
Номер
unknown, С. 19 - 48
Опубликована: Янв. 10, 2025
The
integration
of
artificial
intelligence
(AI)
into
wildlife
conservation
has
revolutionized
methodologies
for
monitoring
species,
enhancing
habitat
management,
and
combating
poaching.
This
chapter
examines
various
AI
applications
that
contribute
to
the
protection
preservation
biodiversity.
Remote
sensing
technologies,
powered
by
machine
learning
algorithms,
assist
in
assessing
health
tracking
changes
over
time.
AI-driven
image
recognition
tools
enable
identification
individual
animals
from
camera
trap
photos,
facilitating
more
accurate
population
estimates
behavioral
studies.
Moreover,
predictive
analytics
play
a
crucial
role
forecasting
human-wildlife
conflicts
informing
proactive
management
strategies.
synthesis
technologies
demonstrates
their
potential
enhance
efforts,
optimize
resource
allocation,
ultimately
foster
effective
initiatives.
ongoing
advancement
this
field
promises
create
innovative
solutions
some
most
pressing
challenges.