Scientific Reports,
Год журнала:
2025,
Номер
15(1)
Опубликована: Апрель 23, 2025
Diabetic
Retinopathy
(DR)
is
a
leading
cause
of
vision
impairment
globally,
necessitating
regular
screenings
to
prevent
its
progression
severe
stages.
Manual
diagnosis
labor-intensive
and
prone
inaccuracies,
highlighting
the
need
for
automated,
accurate
detection
methods.
This
study
proposes
novel
approach
early
DR
by
integrating
advanced
machine
learning
techniques.
The
proposed
system
employs
three-phase
methodology:
initial
image
preprocessing,
blood
vessel
segmentation
using
Hopfield
Neural
Network
(HNN),
feature
extraction
through
an
Attention
Mechanism-based
Capsule
(AM-CapsuleNet).
features
are
optimized
Taylor-based
African
Vulture
Optimization
Algorithm
(AVOA)
classified
Bilinear
Convolutional
(BCAN).
To
enhance
classification
accuracy,
introduces
hybrid
Electric
Fish
Arithmetic
(EFAOA),
which
refines
exploration
phase,
ensuring
rapid
convergence.
model
was
evaluated
on
balanced
dataset
from
APTOS
2019
Blindness
Detection
challenge,
demonstrating
superior
performance
in
terms
accuracy
efficiency.
offers
robust
solution
DR,
potentially
improving
patient
outcomes
timely
precise
diagnosis.
Microarray
technology
has
become
a
vital
tool
in
cardiovascular
research,
enabling
the
simultaneous
analysis
of
thousands
gene
expressions.
This
capability
provides
robust
foundation
for
heart
disease
classification
and
biomarker
discovery.
However,
high
dimensionality,
noise,
sparsity
microarray
data
present
significant
challenges
effective
analysis.
Gene
selection,
which
aims
to
identify
most
relevant
subset
genes,
is
crucial
preprocessing
step
improving
accuracy,
reducing
computational
complexity,
enhancing
biological
interpretability.
Traditional
selection
methods
often
fall
short
capturing
complex,
nonlinear
interactions
among
limiting
their
effectiveness
tasks.
In
this
study,
we
propose
novel
framework
that
leverages
deep
neural
networks
(DNNs)
optimizing
using
data.
DNNs,
known
ability
model
patterns,
are
integrated
with
feature
techniques
address
high-dimensional
The
proposed
method,
DeepGeneNet
(DGN),
combines
DNN-based
into
unified
framework,
ensuring
performance
meaningful
insights
underlying
mechanisms.
Additionally,
incorporates
hyperparameter
optimization
innovative
U-Net
segmentation
further
enhance
accuracy.
These
optimizations
enable
DGN
deliver
scalable
results,
outperforming
traditional
both
predictive
accuracy
Experimental
results
demonstrate
approach
significantly
improves
compared
other
methods.
By
focusing
on
interplay
between
learning,
work
advances
field
genomics,
providing
interpretable
future
applications.
Heart
disease
remains
a
significant
health
threat
due
to
its
high
mortality
rate
and
increasing
prevalence.
Early
prediction
using
basic
physical
markers
from
routine
exams
is
crucial
for
timely
diagnosis
intervention.
However,
manual
analysis
of
large
datasets
can
be
labor-intensive
error-prone.
Our
goal
rapidly
reliably
anticipate
cardiac
variety
body
signs.
This
research
presents
unique
model
heart
prediction.
We
provide
system
predicting
that
blends
the
deep
convolutional
neural
network
with
feature
selection
technique
based
on
LinearSVC.
integrated
method
selects
subset
characteristics
are
strongly
linked
disease.
feed
these
features
into
conventual
we
constructed.
Also
improve
speed
predictor
avoid
gradient
varnishing
or
explosion,
network's
hyperparameters
were
tuned
random
search
algorithm.
The
proposed
was
evaluated
UCI
MIT
datasets.
number
indicators,
such
as
accuracy,
recall,
precision,
F1
score.
results
demonstrate
our
attains
accuracy
rates
98.16%,
98.2%,
95.38%,
97.84%
in
dataset,
an
average
MCC
score
90%.
These
affirm
efficacy
reliability
predict
Life,
Год журнала:
2025,
Номер
15(4), С. 654 - 654
Опубликована: Апрель 16, 2025
Artificial
intelligence
is
rapidly
transforming
quality
assurance
in
healthcare,
driving
advancements
diagnostics,
surgery,
and
patient
care.
This
review
presents
a
comprehensive
analysis
of
artificial
integration—particularly
convolutional
recurrent
neural
networks—across
key
clinical
domains,
significantly
enhancing
diagnostic
accuracy,
surgical
performance,
pathology
evaluation.
intelligence-based
approaches
have
demonstrated
clear
superiority
over
conventional
methods:
networks
achieved
91.56%
accuracy
scanner
fault
detection,
surpassing
manual
inspections;
endoscopic
lesion
detection
sensitivity
rose
from
2.3%
to
6.1%
with
assistance;
gastric
cancer
invasion
depth
classification
reached
89.16%
outperforming
human
endoscopists
by
17.25%.
In
pathology,
93.2%
identifying
out-of-focus
regions
an
F1
score
0.94
lymphocyte
quantification,
promoting
faster
more
reliable
diagnostics.
Similarly,
improved
workflow
recognition
81%
exceeded
95%
skill
assessment
classification.
Beyond
traditional
diagnostics
support,
AI-powered
wearable
sensors,
drug
delivery
systems,
biointegrated
devices
are
advancing
personalized
treatment
optimizing
physiological
monitoring,
automating
care
protocols,
therapeutic
precision.
Despite
these
achievements,
challenges
remain
areas
such
as
data
standardization,
ethical
governance,
model
generalizability.
Overall,
the
findings
underscore
intelligence’s
potential
outperform
techniques
across
multiple
parameters,
emphasizing
need
for
continued
development,
rigorous
validation,
interdisciplinary
collaboration
fully
realize
its
role
precision
medicine
safety.
Polymers for Advanced Technologies,
Год журнала:
2025,
Номер
36(4)
Опубликована: Апрель 1, 2025
ABSTRACT
The
growing
demand
for
self‐powered
wearable
electronic
devices
in
healthcare,
fitness,
and
entertainment
has
driven
significant
advancements
energy
harvesting
technologies.
This
review
explores
the
latest
progress
mechanisms
that
enable
sustainable
autonomous
devices,
with
a
particular
emphasis
on
role
of
polymers
their
development.
Polymers
offer
unique
combination
mechanical
flexibility,
biocompatibility,
lightweight
properties,
making
them
ideal
applications.
systematically
categorizes
major
technologies
into
three
primary
mechanisms:
thermoelectric
generators
(TEGs),
piezoelectric
harvesters
(PEHs),
triboelectric
nanogenerators
(TENGs).
Each
section
provides
an
in‐depth
discussion
working
principles,
material
innovations,
fabrication
techniques,
applications
these
systems.
Beyond
fundamental
mechanisms,
discusses
hybrid
systems
integrate
multiple
sources
to
maximize
power
generation
ensure
continuous
device
operation.
storage
technologies,
such
as
flexible
supercapacitors
micro‐batteries,
is
also
highlighted
address
intermittency
challenges
ambient
sources.
Despite
progress,
remain
improving
conversion
efficiency,
enhancing
durability,
optimizing
system
integration
real‐world
identifies
key
research
directions
overcoming
challenges,
including
advanced
materials
engineering,
miniaturization
artificial
intelligence‐driven
management
strategies.
findings
presented
this
provide
valuable
insights
development
next‐generation
paving
way
efficient
electronics
seamlessly
daily
life.
Scientific Reports,
Год журнала:
2025,
Номер
15(1)
Опубликована: Апрель 23, 2025
Diabetic
Retinopathy
(DR)
is
a
leading
cause
of
vision
impairment
globally,
necessitating
regular
screenings
to
prevent
its
progression
severe
stages.
Manual
diagnosis
labor-intensive
and
prone
inaccuracies,
highlighting
the
need
for
automated,
accurate
detection
methods.
This
study
proposes
novel
approach
early
DR
by
integrating
advanced
machine
learning
techniques.
The
proposed
system
employs
three-phase
methodology:
initial
image
preprocessing,
blood
vessel
segmentation
using
Hopfield
Neural
Network
(HNN),
feature
extraction
through
an
Attention
Mechanism-based
Capsule
(AM-CapsuleNet).
features
are
optimized
Taylor-based
African
Vulture
Optimization
Algorithm
(AVOA)
classified
Bilinear
Convolutional
(BCAN).
To
enhance
classification
accuracy,
introduces
hybrid
Electric
Fish
Arithmetic
(EFAOA),
which
refines
exploration
phase,
ensuring
rapid
convergence.
model
was
evaluated
on
balanced
dataset
from
APTOS
2019
Blindness
Detection
challenge,
demonstrating
superior
performance
in
terms
accuracy
efficiency.
offers
robust
solution
DR,
potentially
improving
patient
outcomes
timely
precise
diagnosis.
