Federated and Centralized Machine Learning for Cell Segmentation: A Comparative Analysis
Electronics,
Год журнала:
2025,
Номер
14(7), С. 1254 - 1254
Опубликована: Март 22, 2025
The
automatic
segmentation
of
cell
images
plays
a
critical
role
in
medicine
and
biology,
as
it
enables
faster
more
accurate
analysis
diagnosis.
Traditional
machine
learning
faces
challenges
since
requires
transferring
sensitive
data
from
laboratories
to
the
cloud,
with
possible
risks
limitations
due
patients’
privacy,
data-sharing
regulations,
or
laboratory
privacy
guidelines.
Federated
addresses
issues
by
introducing
decentralized
approach
that
removes
need
for
laboratories’
sharing.
task
is
divided
among
participating
clients,
each
training
global
model
situated
on
cloud
its
local
dataset.
This
guarantees
only
transmitting
updated
weights
cloud.
In
this
study,
centralized
compared
federated
one,
demonstrating
they
achieve
similar
performances.
Stemming
benchmarking
available
models,
Cellpose,
having
shown
better
recall
precision
(F1=0.84)
than
U-Net
(F1=0.50)
StarDist
(F1=0.12),
was
used
baseline
testbench
implementation.
results
show
both
binary
multi-class
metrics
remain
high
when
employing
solution
(F1=0.86)
(F12clients=0.86).
These
were
also
stable
across
an
increasing
number
clients
reduced
samples
(F14clients=0.87,
F116clients=0.86),
proving
effectiveness
central
aggregation
locally
trained
models.
Язык: Английский
Federated Learning in Chemical Engineering: A Tutorial on a Framework for Privacy-Preserving Collaboration across Distributed Data Sources
Industrial & Engineering Chemistry Research,
Год журнала:
2025,
Номер
unknown
Опубликована: Март 27, 2025
Язык: Английский
Federated Learning for Privacy-Preserving Cybersecurity: A Review on Secure Threat Detection
International Journal of Advanced Research in Science Communication and Technology,
Год журнала:
2025,
Номер
unknown, С. 520 - 528
Опубликована: Апрель 12, 2025
Federated
Learning's
(FL)
distributed
threat
detection
technique
is
a
significant
advancement
in
cybersecurity
as
it
preserves
privacy
while
processing
data
decentralized
manner.
Centralized
security
systems
that
rely
on
raw
collection
present
two
major
threats
to
users
because
they
create
regulatory
problems
addition
breaches.
FL
removes
concerns
through
its
model-building
process,
allowing
different
organizations
work
together
without
sharing
private
data.
This
document
investigates
FL's
role
an
analysis
of
malware/ransomware
detection,
IDS
applications,
secure
and
network
traffic
anomaly
detection.
The
paper
explores
effective
privacy-protecting
techniques:
implementations
are
protected
against
Byzantine
backdoor
attacks
using
Secure
Multi-Party
Computation
(SMPC),
Homomorphic
Encryption
(HE),
Differential
Privacy
(DP),
Model
Aggregation.
delivers
advantages
but
encounters
challenges
mainly
related
excessive
communication
demands
well
performance
deterioration
under
adversarial
conditions,
difficulties
with
system
expansion.
research
provides
exhaustive
FL-based
frameworks
discussing
existing
applications
future
developments
for
these
the
need
advanced
methods
improve
dependability
solutions.
Язык: Английский
A Systematic Review on the Combination of VR, IoT and AI Technologies, and Their Integration in Applications
Future Internet,
Год журнала:
2025,
Номер
17(4), С. 163 - 163
Опубликована: Апрель 7, 2025
The
convergence
of
Virtual
Reality
(VR),
Artificial
Intelligence
(AI),
and
the
Internet
Things
(IoT)
offers
transformative
potential
across
numerous
sectors.
However,
existing
studies
often
examine
these
technologies
independently
or
in
limited
pairings,
which
overlooks
synergistic
possibilities
their
combined
usage.
This
systematic
review
adheres
to
PRISMA
guidelines
order
critically
analyze
peer-reviewed
literature
from
highly
recognized
academic
databases
related
intersection
VR,
AI,
IoT,
identify
application
domains,
methodologies,
tools,
key
challenges.
By
focusing
on
real-life
implementations
working
prototypes,
this
highlights
state-of-the-art
advancements
uncovers
gaps
that
hinder
practical
adoption,
such
as
data
collection
issues,
interoperability
barriers,
user
experience
findings
reveal
digital
twins
(DTs),
AIoT
systems,
immersive
XR
environments
are
promising
emerging
(ET),
but
require
further
development
achieve
scalability
real-world
impact,
while
certain
fields
a
amount
research
is
conducted
until
now.
bridges
theory
practice,
providing
targeted
foundation
for
future
interdisciplinary
aimed
at
advancing
practical,
scalable
solutions
domains
healthcare,
smart
cities,
industry,
education,
cultural
heritage,
beyond.
study
found
integration
IoT
holds
significant
various
with
DTs,
showing
applications,
challenges
interoperability,
limitations,
barriers
widespread
adoption.
Язык: Английский
Generative AI for Cybersecurity Applications in Threat Simulation and Defense
Advances in computational intelligence and robotics book series,
Год журнала:
2025,
Номер
unknown, С. 263 - 304
Опубликована: Апрель 23, 2025
The
integration
of
generative
AI
in
cybersecurity
marks
a
transformative
leap
combating
the
growing
complexity
cyber
threats.
This
chapter
examines
models
like
adversarial
networks,
variational
autoencoders,
and
transformers,
showcasing
their
role
threat
simulation,
synthetic
data
generation,
anomaly
detection.
Applications
discussed
include
proactive
defense
testing,
malware
analysis,
intrusion
detection,
highlighting
AI's
ability
to
predict,
detect,
mitigate
sophisticated
attacks.
Emerging
techniques,
such
as
federated
learning
hybrid
models,
promise
further
advancements.
However,
poses
challenges,
including
misuse
vulnerabilities.
Addressing
these
risks
requires
ethical
guidelines,
robust
frameworks,
collaboration.
With
its
predictive
adaptive
potential,
is
reshaping
cybersecurity,
enabling
resilient
intelligent
defenses
for
digital
age.
Язык: Английский
CLDM-MMNNs: Cross-layer defense mechanisms through multi-modal neural networks fusion for end-to-end cybersecurity—Issues, challenges, and future directions
Information Fusion,
Год журнала:
2025,
Номер
unknown, С. 103222 - 103222
Опубликована: Апрель 1, 2025
Язык: Английский
Enhanced Federated Learning Framework for Edge-Enabled Green IoT
Journal of Information Technology and Digital World,
Год журнала:
2025,
Номер
7(1), С. 56 - 67
Опубликована: Март 1, 2025
The
Internet
of
Things
(IoT)
is
rapidly
transforming
industries
by
enabling
seamless
data
collection
and
processing.
However,
the
massive
influx
poses
significant
challenges
in
terms
energy
consumption
privacy.
Federated
Learning
(FL)
has
emerged
as
a
promising
solution,
allowing
distributed
model
training
without
transmitting
raw
data.
This
research
proposes
an
Enhanced
Framework
(EFLF)
for
edge-enabled
green
IoT
that
optimizes
efficiency
while
maintaining
high
accuracy.
proposed
framework
integrates
adaptive
client
selection,
energy-aware
aggregation,
compression
techniques.
Experimental
results
demonstrate
superior
performance
convergence
compared
to
baseline
FL
approaches.
Язык: Английский