Acta medica Lituanica,
Journal Year:
2024,
Volume and Issue:
31(2), P. 294 - 301
Published: Dec. 26, 2024
Artificial
intelligence
models
human
thinking
and
problem-solving
abilities,
allowing
computers
to
make
autonomous
decisions.
There
is
a
lack
of
studies
demonstrating
the
clinical
utility
GPT
Gemin
in
field
toxicology,
which
means
their
level
competence
not
well
understood.
This
study
compares
responses
given
by
GPT-3.5
those
provided
emergency
medicine
residents.
prospective
was
focused
on
toxicology
utilized
widely
recognized
educational
resource
'Tintinalli
Emergency
Medicine:
A
Comprehensive
Study
Guide'
for
Medicine.
set
twenty
questions,
each
with
five
options,
devised
test
knowledge
toxicological
data
as
defined
book.
These
questions
were
then
used
train
ChatGPT
(Generative
Pre-trained
Transformer
3.5)
OpenAI
Gemini
Google
AI
clinic.
The
resulting
answers
meticulously
analyzed.
28
physicians,
35.7%
whom
women,
included
our
study.
comparison
made
between
physician
scores.
While
significant
difference
found
(F=2.368
p<0.001),
no
two
groups
post-hoc
Tukey
test.
mean
score
9.9±0.71,
11.30±1.17
and,
physicians'
9.82±3.70
(Figure
1).
It
clear
that
respond
similarly
topics
just
resident
physicians
do.
Artificial Intelligence Review,
Journal Year:
2024,
Volume and Issue:
57(5)
Published: April 25, 2024
Abstract
Background
Precision
medicine,
targeting
treatments
to
individual
genetic
and
clinical
profiles,
faces
challenges
in
data
collection,
costs,
privacy.
Generative
AI
offers
a
promising
solution
by
creating
realistic,
privacy-preserving
patient
data,
potentially
revolutionizing
patient-centric
healthcare.
Objective
This
review
examines
the
role
of
deep
generative
models
(DGMs)
informatics,
medical
imaging,
bioinformatics,
early
diagnostics,
showcasing
their
impact
on
precision
medicine.
Methods
Adhering
PRISMA
guidelines,
analyzes
studies
from
databases
such
as
Scopus
PubMed,
focusing
AI's
medicine
DGMs'
applications
synthetic
generation.
Results
DGMs,
particularly
Adversarial
Networks
(GANs),
have
improved
generation,
enhancing
accuracy
However,
limitations
exist,
especially
foundation
like
Large
Language
Models
(LLMs)
digital
diagnostics.
Conclusion
Overcoming
scarcity
ensuring
privacy-safe
generation
are
crucial
for
advancing
personalized
Further
development
LLMs
is
essential
improving
diagnostic
precision.
The
application
emerging,
highlighting
need
more
interdisciplinary
research
advance
this
field.
Cureus,
Journal Year:
2024,
Volume and Issue:
unknown
Published: Jan. 14, 2024
Autonomous
robotic
surgery
represents
a
pioneering
field
dedicated
to
the
integration
of
systems
with
varying
degrees
autonomy
for
execution
surgical
procedures.
This
paradigm
shift
is
made
possible
by
progressive
artificial
intelligence
(AI)
and
machine
learning
(ML)
into
realm
interventions.
While
majority
autonomous
remain
in
experimental
phase,
notable
subset
has
successfully
transitioned
clinical
applications.
Noteworthy
procedures,
such
as
venipuncture,
hair
implantations,
intestinal
anastomosis,
total
knee
replacement,
cochlear
implant,
radiosurgery,
knot
tying,
among
others,
exemplify
current
capabilities
systems.
review
endeavors
comprehensively
address
facets
surgery,
commencing
concise
elucidation
fundamental
concepts
traversing
pivotal
milestones
historical
evolution
surgery.
trajectory
underscores
incremental
assimilation
practices.
aims
topics
related
starting
description
going
through
history
that
also
show
gradual
incorporations
It
includes
discussion
key
benefits
risks
this
technology,
robots,
their
limitations,
legal
regulations
governing
usage,
main
ethical
concerns
inherent
nature.
Brain and Behavior,
Journal Year:
2025,
Volume and Issue:
15(2)
Published: Feb. 1, 2025
ABSTRACT
Background
The
relentless
integration
of
Artificial
Intelligence
(AI)
into
neurosurgery
necessitates
a
meticulous
exploration
the
associated
ethical
concerns.
This
systematic
review
focuses
on
synthesizing
empirical
studies,
reviews,
and
opinion
pieces
from
past
decade,
offering
nuanced
understanding
evolving
intersection
between
AI
neurosurgical
ethics.
Materials
Methods
Following
PRISMA
guidelines,
was
conducted
to
identify
studies
addressing
in
neurosurgery,
emphasizing
dimensions.
search
strategy
employed
keywords
related
AI,
Inclusion
criteria
encompassed
analyses
published
last
with
English
language
restriction.
Quality
assessment
using
Joanna
Briggs
Institute
tools
ensured
methodological
rigor.
Results
Eight
key
were
identified,
each
contributing
unique
insights
considerations
neurosurgery.
Findings
highlighted
limitations
technologies,
challenges
data
bias,
transparency,
legal
responsibilities.
emphasized
need
for
responsible
systems,
regulatory
oversight,
transparent
decision‐making
practices.
Conclusions
synthesis
findings
underscores
complexity
Transparent
use,
mitigation
biases
emerged
as
recurring
themes.
calls
establishment
comprehensive
guidelines
ensure
safe
equitable
Ongoing
research,
educational
initiatives,
culture
innovation
are
crucial
navigating
landscape
AI‐driven
advancements
UNSTRUCTURED
Purpose:
This
systematic
review
examines
the
potential
of
ChatGPT
as
a
tool
in
medical
education,
focusing
on
its
role
enhancing
learning
experiences,
student
performance,
and
critical
thinking
skills.
ChatGPT's
integration
aims
to
address
shortage
faculty
resources
create
personalized,
interactive
experiences
for
students.
Methods:
Following
PRISMA
AMSTAR
guidelines,
we
conducted
across
four
databases
(Embase,
PubMed,
Web
Science,
Cochrane
Library)
up
October
2024.
Data
from
seven
studies
various
disciplines
educational
levels
were
included,
analyzed
descriptively,
evaluated
quality.
Results:
Seven
demonstrated
that
ChatGPT-assisted
education
improves
academic
clinical
skills,
SDL
capabilities.
Notably,
students
using
showed
higher
scores
short-term
assessments
final
exams.
4.0,
compared
version
3.5,
provided
enhanced
case
generation
communication
skills
training.
Additionally,
ChatGPT-supported
boosted
students'
SDL,
thinking,
engagement
levels,
while
helping
educators
manage
instructional
workload.
Conclusion:
study
highlights
ChatGPT’s
strong
significantly
self-directed
learning,
thinking.
It
underscores
personalized
supporting
development
essential
competencies.
4.0
outperforms
3.5
with
improved
abilities.
Frontiers in Medicine,
Journal Year:
2025,
Volume and Issue:
11
Published: Jan. 30, 2025
Large
Language
Models
(LLMs)
like
ChatGPT,
Gemini,
and
Claude
gain
traction
in
healthcare
simulation;
this
paper
offers
simulationists
a
practical
guide
to
effective
prompt
design.
Grounded
structured
literature
review
iterative
testing,
proposes
best
practices
for
developing
calibrated
prompts,
explores
various
types
techniques
with
use
cases,
addresses
the
challenges,
including
ethical
considerations
using
LLMs
simulation.
This
helps
bridge
knowledge
gap
on
LLM
simulation-based
education,
offering
tailored
guidance
Examples
were
created
through
testing
ensure
alignment
simulation
objectives,
covering
cases
such
as
clinical
scenario
development,
OSCE
station
creation,
simulated
person
scripting,
debriefing
facilitation.
These
provide
easy-to-apply
methods
enhance
realism,
engagement,
educational
simulations.
Key
challenges
associated
integration,
bias,
privacy
concerns,
hallucinations,
lack
of
transparency,
need
robust
oversight
evaluation,
are
discussed
alongside
unique
education.
Recommendations
provided
help
craft
prompts
that
align
objectives
while
mitigating
these
challenges.
By
insights,
contributes
valuable,
timely
seeking
leverage
generative
AI’s
capabilities
education
responsibly.
Cureus,
Journal Year:
2025,
Volume and Issue:
unknown
Published: March 18, 2025
This
case
study
explores
the
integration
of
Viz
ICH
Plus,
an
AI-powered
intracerebral
hemorrhage
(ICH)
detection
system,
into
a
centralized
program
called
Neuroemergencies
Management
and
Transfer
(NEMAT)
large
urban
healthcare
system.
The
highlights
how
Plus
promptly
identified
right
parieto-occipital
hematoma
in
patient
presenting
with
headache,
resulting
marked
reduction
interhospital
transfer
(IHT)
time.
underwent
successful
supratentorial
craniotomy
for
evacuation
demonstrated
significant
cognitive
physical
improvement
over
following
year.
reduced
IHT
time
from
approximately
200
to
101
minutes,
expediting
access
definitive
care
improving
outcomes.
Standard
radiology
review
scan
communication
results
could
have
added
additional
delays
transferring
this
receive
care.
illustrates
substantial
potential
AI
transform
stroke
by
optimizing
response
times
facilitating
timely
interventions.