Informatics,
Journal Year:
2025,
Volume and Issue:
12(1), P. 9 - 9
Published: Jan. 17, 2025
The
rapid
advancement
of
large
language
models
like
ChatGPT
has
significantly
impacted
natural
processing,
expanding
its
applications
across
various
fields,
including
healthcare.
However,
there
remains
a
significant
gap
in
understanding
the
consistency
and
reliability
ChatGPT’s
performance
different
medical
domains.
We
conducted
this
systematic
review
according
to
an
LLM-assisted
PRISMA
setup.
high-recall
search
term
“ChatGPT”
yielded
1101
articles
from
2023
onwards.
Through
dual-phase
screening
process,
initially
automated
via
subsequently
manually
by
human
reviewers,
128
studies
were
included.
covered
range
specialties,
focusing
on
diagnosis,
disease
management,
patient
education.
assessment
metrics
varied,
but
most
compared
accuracy
against
evaluations
clinicians
or
reliable
references.
In
several
areas,
demonstrated
high
accuracy,
underscoring
effectiveness.
some
contexts
revealed
lower
accuracy.
mixed
outcomes
domains
emphasize
challenges
opportunities
integrating
AI
into
certain
areas
suggests
that
substantial
utility,
yet
inconsistent
all
indicates
need
for
ongoing
evaluation
refinement.
This
highlights
potential
improve
healthcare
delivery
alongside
necessity
continued
research
ensure
reliability.
Respiratory Research,
Journal Year:
2025,
Volume and Issue:
26(1)
Published: Feb. 12, 2025
Tuberculous
pleural
effusion
(TPE)
is
a
challenging
extrapulmonary
manifestation
of
tuberculosis,
with
traditional
diagnostic
methods
often
involving
invasive
surgery
and
being
time-consuming.
While
various
machine
learning
statistical
models
have
been
proposed
for
TPE
diagnosis,
these
are
typically
limited
by
complexities
in
data
processing
difficulties
feature
integration.
Therefore,
this
study
aims
to
develop
model
using
ChatGPT-4,
large
language
(LLM),
compare
its
performance
logistic
regression
models.
By
highlighting
the
advantages
LLMs
handling
complex
clinical
data,
identifying
interrelationships
between
features,
improving
accuracy,
seeks
provide
more
efficient
precise
solution
early
diagnosis
TPE.
We
conducted
cross-sectional
study,
collecting
from
109
54
non-TPE
patients
analysis,
selecting
73
features
over
600
initial
variables.
The
LLM
was
compared
(k-Nearest
Neighbors,
Random
Forest,
Support
Vector
Machines)
metrics
like
area
under
curve
(AUC),
F1
score,
sensitivity,
specificity.
showed
comparable
models,
outperforming
specificity,
overall
accuracy.
Key
such
as
adenosine
deaminase
(ADA)
levels
monocyte
percentage
were
effectively
integrated
into
model.
also
developed
Python
package
(
https://pypi.org/project/tpeai/
)
rapid
based
on
data.
LLM-based
offers
non-surgical,
accurate,
cost-effective
method
diagnosis.
provides
user-friendly
tool
clinicians,
potential
broader
use.
Further
validation
larger
datasets
needed
optimize
application.
Dental Press Journal of Orthodontics,
Journal Year:
2023,
Volume and Issue:
28(5)
Published: Jan. 1, 2023
ABSTRACT
Introduction:
Artificial
Intelligence
(AI)
is
a
tool
that
already
part
of
our
reality,
and
this
an
opportunity
to
understand
how
it
can
be
useful
in
interacting
with
patients
providing
valuable
information
about
orthodontics.
Objective:
This
study
evaluated
the
accuracy
ChatGPT
accurate
quality
answer
questions
on
Clear
aligners,
Temporary
anchorage
devices
Digital
imaging
Methods:
forty-five
answers
were
generated
by
4.0,
analyzed
separately
five
orthodontists.
The
evaluators
independently
rated
provided
Likert
scale,
which
higher
scores
indicated
greater
(1
=
very
poor;
2
3
acceptable;
4
good;
5
good).
Kruskal-Wallis
H
test
(p<
0.05)
post-hoc
pairwise
comparisons
Bonferroni
correction
performed.
Results:
From
225
evaluations
different
evaluators,
11
(4.9%)
considered
as
poor,
(1.8%)
15
(6.7%)
acceptable.
majority
good
[34
(15,1%)]
[161
(71.6%)].
Regarding
evaluators’
scores,
slight
agreement
was
perceived,
Fleiss’s
Kappa
equal
0.004.
Conclusions:
has
proven
effective
related
clear
temporary
devices,
digital
within
context
interest
Journal of Pediatric Orthopaedics,
Journal Year:
2024,
Volume and Issue:
44(7), P. e592 - e597
Published: April 30, 2024
Objective:
Chat
generative
pre-trained
transformer
(ChatGPT)
has
garnered
attention
in
health
care
for
its
potential
to
reshape
patient
interactions.
As
patients
increasingly
rely
on
artificial
intelligence
platforms,
concerns
about
information
accuracy
arise.
In-toeing,
a
common
lower
extremity
variation,
often
leads
pediatric
orthopaedic
referrals
despite
observation
being
the
primary
treatment.
Our
study
aims
assess
ChatGPT’s
responses
in-toeing
questions,
contributing
discussions
innovation
and
technology
education.
Methods:
We
compiled
list
of
34
questions
from
“Frequently
Asked
Questions”
sections
9
care–affiliated
websites,
identifying
25
as
most
encountered.
On
January
17,
2024,
we
queried
ChatGPT
3.5
separate
sessions
recorded
responses.
These
were
posed
again
21,
reproducibility.
Two
surgeons
evaluated
using
scale
“excellent
(no
clarification)”
“unsatisfactory
(substantial
clarification).”
Average
ratings
used
when
evaluators’
grades
within
one
level
each
other.
In
discordant
cases,
senior
author
provided
decisive
rating.
Results:
found
46%
“excellent”
44%
“satisfactory
(minimal
addition,
8%
cases
(moderate
2%
“unsatisfactory.”
Questions
had
appropriate
readability,
with
an
average
Flesch-Kincaid
Grade
Level
4.9
(±2.1).
However,
at
collegiate
level,
averaging
12.7
(±1.4).
No
significant
differences
observed
between
question
topics.
Furthermore,
exhibited
moderate
consistency
after
repeated
queries,
evidenced
by
Spearman
rho
coefficient
0.55
(
P
=
0.005).
The
chatbot
appropriately
described
normal
or
spontaneously
resolving
62%
consistently
recommended
evaluation
provider
100%.
Conclusion:
presented
serviceable,
though
not
perfect,
representation
diagnosis
management
while
demonstrating
reproducibility
utility
could
be
enhanced
improving
readability
incorporating
evidence-based
guidelines.
Evidence:
IV—diagnostic.
Asia-Pacific Journal of Ophthalmology,
Journal Year:
2024,
Volume and Issue:
13(4), P. 100085 - 100085
Published: July 1, 2024
Large
language
models
(LLMs),
a
natural
processing
technology
based
on
deep
learning,
are
currently
in
the
spotlight.
These
closely
mimic
comprehension
and
generation.
Their
evolution
has
undergone
several
waves
of
innovation
similar
to
convolutional
neural
networks.
The
transformer
architecture
advancement
generative
artificial
intelligence
marks
monumental
leap
beyond
early-stage
pattern
recognition
via
supervised
learning.
With
expansion
parameters
training
data
(terabytes),
LLMs
unveil
remarkable
human
interactivity,
encompassing
capabilities
such
as
memory
retention
comprehension.
advances
make
particularly
well-suited
for
roles
healthcare
communication
between
medical
practitioners
patients.
In
this
comprehensive
review,
we
discuss
trajectory
their
potential
implications
clinicians
For
clinicians,
can
be
used
automated
documentation,
given
better
inputs
extensive
validation,
may
able
autonomously
diagnose
treat
future.
patient
care,
triage
suggestions,
summarization
documents,
explanation
patient's
condition,
customizing
education
materials
tailored
level.
limitations
possible
solutions
real-world
use
also
presented.
Given
rapid
advancements
area,
review
attempts
briefly
cover
many
that
play
ophthalmic
space,
with
focus
improving
quality
delivery.
International Dental Journal,
Journal Year:
2024,
Volume and Issue:
unknown
Published: Aug. 1, 2024
Given
the
increasing
interest
in
using
large
language
models
(LLMs)
for
self-diagnosis,
this
study
aimed
to
evaluate
comprehensiveness
of
two
prominent
LLMs,
ChatGPT-3.5
and
ChatGPT-4,
addressing
common
queries
related
gingival
endodontic
health
across
different
contexts
query
types.
British Journal of Ophthalmology,
Journal Year:
2023,
Volume and Issue:
108(10), P. 1362 - 1370
Published: Dec. 11, 2023
Large
language
models
(LLMs)
are
fast
emerging
as
potent
tools
in
healthcare,
including
ophthalmology.
This
systematic
review
offers
a
twofold
contribution:
it
summarises
current
trends
ophthalmology-related
LLM
research
and
projects
future
directions
for
this
burgeoning
field.
