npj Digital Medicine,
Journal Year:
2025,
Volume and Issue:
8(1)
Published: May 14, 2025
Summary
Anonymized
biomedical
data
sharing
faces
several
challenges.
This
systematic
review
analyzes
1084
PubMed-indexed
studies
(2018–2022)
using
anonymized
to
quantify
usage
trends
across
geographic,
regulatory,
and
cultural
regions
identify
effective
approaches
inform
implementation
agendas.
We
identified
a
significant
yearly
increase
in
such
with
slope
of
2.16
articles
per
100,000
when
normalized
against
the
total
number
(
p
=
0.021).
Most
used
from
US,
UK,
Australia
(78.2%).
trend
remained
by
country-specific
research
output.
Cross-border
was
rare
(10.5%
studies).
twelve
common
sources,
primarily
US
(seven)
UK
(three),
including
commercial
public
entities
(five).
The
prevalence
anonymization
suggests
their
practices
could
guide
broader
adoption.
Rare
cross-border
differences
between
countries
comparable
regulations
underscore
need
for
global
standards.
New England Journal of Medicine,
Journal Year:
2023,
Volume and Issue:
388(21), P. 1981 - 1990
Published: May 24, 2023
The
authors
examine
the
advantages
and
limitations
of
current
clinical
radiologic
AI
systems,
new
workflows,
potential
effect
generative
large
multimodal
foundation
models.
Scientific Data,
Journal Year:
2023,
Volume and Issue:
10(1)
Published: June 24, 2023
Sharing
healthcare
data
is
increasingly
essential
for
developing
data-driven
improvements
in
patient
care
at
the
Intensive
Care
Unit
(ICU).
However,
it
also
very
challenging
under
strict
privacy
legislation
of
European
Union
(EU).
Therefore,
we
explored
four
successful
open
ICU
databases
to
determine
how
can
be
shared
appropriately
EU.
A
questionnaire
was
constructed
based
on
Delphi
method.
Then,
follow-up
questions
were
discussed
with
experts
from
databases.
These
encountered
similar
challenges
and
regarded
ethical
legal
aspects
most
challenging.
Based
approaches
databases,
expert
opinion,
literature
research,
outline
distinct
openly
sharing
data,
each
varying
implications
regarding
security,
ease
use,
sustainability,
implementability.
Ultimately,
formulate
seven
recommendations
guide
future
initiatives
improve
advance
healthcare.
Journal of Cancer Research and Clinical Oncology,
Journal Year:
2023,
Volume and Issue:
149(10), P. 7997 - 8006
Published: March 15, 2023
Artificial
intelligence
(AI)
is
influencing
our
society
on
many
levels
and
has
broad
implications
for
the
future
practice
of
hematology
oncology.
However,
medical
professionals
researchers,
it
often
remains
unclear
what
AI
can
cannot
do,
are
promising
areas
a
sensible
application
in
Finally,
limits
perils
using
oncology
not
obvious
to
healthcare
professionals.
BMC Medical Informatics and Decision Making,
Journal Year:
2025,
Volume and Issue:
25(1)
Published: Feb. 5, 2025
Abstract
Background
Artificial
intelligence
(AI)-based
systems
are
being
rapidly
integrated
into
the
fields
of
health
and
social
care.
Although
such
can
substantially
improve
provision
care,
diverse
marginalized
populations
often
incorrectly
or
insufficiently
represented
within
these
systems.
This
review
aims
to
assess
influence
AI
on
care
among
populations,
particularly
with
regard
issues
related
inclusivity
regulatory
concerns.
Methods
We
followed
Preferred
Reporting
Items
for
Systematic
Reviews
Meta-Analyses
guidelines.
Six
leading
databases
were
searched,
129
articles
selected
this
in
line
predefined
eligibility
criteria.
Results
research
revealed
disparities
outcomes,
accessibility,
representation
groups
due
biased
data
sources
a
lack
training
datasets,
which
potentially
exacerbate
inequalities
delivery
communities.
Conclusion
development
practices,
legal
frameworks,
policies
must
be
reformulated
ensure
that
is
applied
an
equitable
manner.
A
holistic
approach
used
address
disparities,
enforce
effective
regulations,
safeguard
privacy,
promote
inclusion
equity,
emphasize
rigorous
validation.
Big Data and Cognitive Computing,
Journal Year:
2023,
Volume and Issue:
7(1), P. 32 - 32
Published: Feb. 9, 2023
Preeclampsia
is
one
of
the
illnesses
associated
with
placental
dysfunction
and
pregnancy-induced
hypertension,
which
appears
after
first
20
weeks
pregnancy
marked
by
proteinuria
hypertension.
It
can
affect
pregnant
women
limit
fetal
growth,
resulting
in
low
birth
weights,
a
risk
factor
for
neonatal
mortality.
Approximately
10%
pregnancies
worldwide
are
affected
hypertensive
disorders
during
pregnancy.
In
this
review,
we
discuss
machine
learning
deep
methods
preeclampsia
prediction
that
were
published
between
2018
2022.
Many
models
have
been
created
using
variety
data
types,
including
demographic
clinical
data.
We
determined
techniques
successfully
predicted
preeclampsia.
The
used
most
random
forest,
support
vector
machine,
artificial
neural
network
(ANN).
addition,
prospects
challenges
discussed
to
boost
research
on
intelligence
systems,
allowing
academics
practitioners
improve
their
advance
automated
prediction.
medRxiv (Cold Spring Harbor Laboratory),
Journal Year:
2024,
Volume and Issue:
unknown
Published: Jan. 23, 2024
Abstract
Introduction
The
Brazilian
Multilabel
Ophthalmological
Dataset
(BRSET)
addresses
the
scarcity
of
publicly
available
ophthalmological
datasets
in
Latin
America.
BRSET
comprises
16,266
color
fundus
retinal
photos
from
8,524
patients,
aiming
to
enhance
data
representativeness,
serving
as
a
research
and
teaching
tool.
It
contains
sociodemographic
information,
enabling
investigations
into
differential
model
performance
across
demographic
groups.
Methods
Data
three
São
Paulo
outpatient
centers
yielded
medical
information
electronic
records,
including
nationality,
age,
sex,
clinical
history,
insulin
use,
duration
diabetes
diagnosis.
A
specialist
labeled
images
for
anatomical
features
(optic
disc,
blood
vessels,
macula),
quality
control
(focus,
illumination,
image
field,
artifacts),
pathologies
(e.g.,
diabetic
retinopathy).
Diabetic
retinopathy
was
graded
using
International
Clinic
Retinopathy
Scottish
Grading.
Validation
used
Dino
V2
Base
feature
extraction,
with
70%
training
30%
testing
subsets.
Support
Vector
Machines
(SVM)
Logistic
Regression
(LR)
were
employed
weighted
training.
Performance
metrics
included
area
under
receiver
operating
curve
(AUC)
Macro
F1-score.
Results
65.1%
Canon
CR2
34.9%
Nikon
NF5050
images.
61.8%
patients
are
female,
average
age
is
57.6
years.
affected
15.8%
spectrum
disease
severity.
Anatomically,
20.2%
showed
abnormal
optic
discs,
4.9%
28.8%
macula.
Models
trained
on
prediction
tasks:
“diabetes
diagnosis”;
“sex
classification”;
“diabetic
diagnosis”.
Discussion
first
multilabel
dataset
Brazil
provides
an
opportunity
investigating
biases
by
evaluating
tasks
demonstrates
value
external
validation
computer
vision
learners
America
locally
relevant
sources.
Author
Summary
In
low-resource
settings,
access
open
crucial
research.
Regions
such
often
face
underrepresentation,
resulting
health
inequities.
To
diverse
these
areas,
especially
America,
we
introduce
means
alleviate
AI
Comprising
integrates
empowering
researchers
investigate
groups
diseases.
extracted
centers,
includes
demographics,
features,
control,
like
retinopathy.
performed
set
selected
tasks,
diagnosis,
sex
classification,
BRSET’s
inclusion
experiment
underscores
its
potential
efficacy
classification
objectives
patient
groups,
providing
insights
underrepresented
regions.
Journal of Medical Internet Research,
Journal Year:
2023,
Volume and Issue:
25, P. e43333 - e43333
Published: June 22, 2023
Artificial
Intelligence
(AI)
represents
a
significant
milestone
in
health
care's
digital
transformation.
However,
traditional
care
education
and
training
often
lack
competencies.
To
promote
safe
effective
AI
implementation,
professionals
must
acquire
basic
knowledge
of
machine
learning
neural
networks,
critical
evaluation
data
sets,
integration
within
clinical
workflows,
bias
control,
human-machine
interaction
settings.
Additionally,
they
should
understand
the
legal
ethical
aspects
impact
adoption.
Misconceptions
fears
about
systems
could
jeopardize
its
real-life
implementation.
there
are
multiple
barriers
to
promoting
electronic
literacy,
including
time
constraints,
overburdened
curricula,
shortage
capacitated
professionals.
overcome
these
challenges,
partnerships
among
developers,
professional
societies,
academia
essential.
Integrating
specialists
from
different
backgrounds,
specialists,
lawyers,
social
scientists,
can
significantly
contribute
combating
illiteracy
implementation
care.
Allergy,
Journal Year:
2023,
Volume and Issue:
78(10), P. 2623 - 2643
Published: Aug. 16, 2023
Abstract
The
field
of
medicine
is
witnessing
an
exponential
growth
interest
in
artificial
intelligence
(AI),
which
enables
new
research
questions
and
the
analysis
larger
types
data.
Nevertheless,
applications
that
go
beyond
proof
concepts
deliver
clinical
value
remain
rare,
especially
allergy.
This
narrative
review
provides
a
fundamental
understanding
core
AI
critically
discusses
its
limitations
open
challenges,
such
as
data
availability
bias,
along
with
potential
directions
to
surmount
them.
We
provide
conceptual
framework
structure
within
this
discuss
forefront
case
examples.
Most
these
machine
learning
allergy
concern
supervised
unsupervised
clustering,
strong
emphasis
on
diagnosis
subtyping.
A
perspective
shared
guidelines
for
good
practice
guide
readers
applying
it
effectively
safely,
prospects
advancement
initiatives
increase
impact.
anticipate
can
further
deepen
our
knowledge
disease
mechanisms
contribute
precision
