Indian journal of radiology and imaging - new series/Indian journal of radiology and imaging/Indian Journal of Radiology & Imaging,
Journal Year:
2024,
Volume and Issue:
unknown
Published: Nov. 4, 2024
Abstract
Background
Radiology
is
critical
for
diagnosis
and
patient
care,
relying
heavily
on
accurate
image
interpretation.
Recent
advancements
in
artificial
intelligence
(AI)
natural
language
processing
(NLP)
have
raised
interest
the
potential
of
AI
models
to
support
radiologists,
although
robust
research
performance
this
field
still
emerging.
Objective
This
study
aimed
assess
efficacy
ChatGPT-4
answering
radiological
anatomy
questions
similar
those
Fellowship
Royal
College
Radiologists
(FRCR)
Part
1
Anatomy
examination.
Methods
We
used
100
mock
from
a
free
Web
site
patterned
after
FRCR
was
tested
under
two
conditions:
with
without
context
regarding
examination
instructions
question
format.
The
main
query
posed
was:
“Identify
structure
indicated
by
arrow(s).”
Responses
were
evaluated
against
correct
answers,
expert
radiologists
(>5
30
years
experience
radiology
diagnostics
academics)
rated
explanation
answers.
calculated
four
scores:
correctness,
sidedness,
modality
identification,
approximation.
latter
considers
partial
correctness
if
identified
present
but
not
focus
question.
Results
Both
testing
conditions
saw
underperform,
scores
4
7.5%
no
context,
respectively.
However,
it
imaging
100%
accuracy.
model
scored
over
50%
approximation
metric,
where
structures
arrow.
struggled
identifying
side
structure,
scoring
approximately
42
40%
settings,
Only
32%
responses
across
settings.
Conclusion
Despite
its
ability
correctly
recognize
modality,
has
significant
limitations
interpreting
normal
anatomy.
indicates
necessity
enhanced
training
better
interpret
abnormal
images.
Identifying
images
also
remains
challenge
ChatGPT-4.
BACKGROUND
Polycystic
ovary
syndrome
(PCOS)
is
a
prevalent
condition
requiring
effective
patient
education,
particularly
in
China.
Large
language
models
(LLMs)
present
promising
avenue
for
this.
This
two-phase
study
evaluates
six
LLMs
educating
Chinese
patients
about
PCOS.
It
assesses
their
capabilities
answering
questions,
interpreting
ultrasound
images,
and
providing
instructions
within
real-world
clinical
setting
OBJECTIVE
systematically
evaluated
gigantic
models—Gemini
2.0
Pro,
OpenAI
o1,
ChatGPT-4o,
ChatGPT-4,
ERINE
4.0,
GLM-4—for
use
gynecological
medicine.
assessed
performance
several
areas:
questions
from
the
Gynecology
Qualification
Examination,
understanding
coping
with
polycystic
cases,
writing
instructions,
helping
to
solve
problems.
METHODS
A
two-step
evaluation
method
was
used.
Primarily,
they
tested
frameworks
on
136
exam
36
images.
They
then
compared
results
those
of
medical
students
residents.
Six
gynecologists
framework's
responses
23
PCOS-related
using
Likert
scale,
readability
tool
used
review
content
objectively.
In
following
process,
40
PCOS
two
central
systems,
Gemini
Pro
o1.
them
terms
satisfaction,
text
readability,
professional
evaluation.
RESULTS
During
initial
phase
testing,
o1
demonstrated
impressive
accuracy
specialist
achieving
rates
93.63%
92.40%,
respectively.
Additionally,
image
diagnostic
tasks
noteworthy,
an
69.44%
reaching
53.70%.
Regarding
response
significantly
outperformed
other
accuracy,
completeness,
practicality,
safety.
However,
its
were
notably
more
complex
(average
score
13.98,
p
=
0.003).
The
second-phase
revealed
that
excelled
(patient
rating
3.45,
<
0.01;
physician
3.35,
0.03),
surpassing
2.65,
2.90).
slightly
lagged
behind
completeness
(3.05
vs.
3.50,
0.04).
CONCLUSIONS
reveals
large
have
considerable
potential
address
issues
faced
by
PCOS,
which
are
capable
accurate
comprehensive
responses.
Nevertheless,
it
still
needs
be
strengthened
so
can
balance
clarity
comprehensiveness.
addition,
big
besides
analyzing
especially
ability
handle
regulation
categories,
improved
meet
practice.
CLINICALTRIAL
None
Frontiers in Endocrinology,
Journal Year:
2025,
Volume and Issue:
16
Published: Feb. 28, 2025
Background
Increasing
numbers
of
cytologically
indeterminate
thyroid
nodules
(ITNs)
present
challenges
for
preoperative
diagnosis,
often
leading
to
unnecessary
diagnostic
surgical
procedures
that
prove
benign.
Research
in
ultrasound
radiomics
and
genomic
testing
leverages
high-throughput
data
image
or
sequence
algorithms
establish
assisted
models
panels
ITN
diagnosis.
Many
now
demonstrate
accuracy
above
80%
sensitivity
over
90%,
surpassing
the
performance
less
experienced
radiologists
and,
some
cases,
matching
radiologists.
Molecular
have
helped
clinicians
achieve
accurate
diagnoses
ITNs,
preventing
42%–61%
patients
with
benign
nodules.
Objective
In
this
review,
we
examined
studies
on
molecular
cytological
ITNs
conducted
past
5
years,
aiming
provide
insights
researchers
focused
improving
Conclusion
Radiomics
enhanced
before
surgery
reduced
patients.
Deleted Journal,
Journal Year:
2025,
Volume and Issue:
unknown
Published: March 3, 2025
The
study
evaluates
the
appropriateness
and
reliability
of
thyroid
nodule
cancer
risk
assessment
recommendations
provided
by
large
language
models
(LLMs)
ChatGPT,
Gemini,
Claude
in
alignment
with
clinical
guidelines
from
American
Thyroid
Association
(ATA)
National
Comprehensive
Cancer
Network
(NCCN).
A
team
comprising
a
medical
imaging
informatics
specialist
two
radiologists
developed
24
clinically
relevant
questions
based
on
ATA
NCCN
guidelines.
readability
AI-generated
responses
was
evaluated
using
Readability
Scoring
System.
total
322
training
or
practice
United
States,
recruited
via
Amazon
Mechanical
Turk,
assessed
AI
responses.
Quantitative
analysis
SPSS
measured
recommendations,
while
qualitative
feedback
analyzed
through
Dedoose.
compared
performance
three
providing
appropriate
recommendations.
Paired
samples
t-tests
showed
no
statistically
significant
differences
overall
among
models.
achieved
highest
mean
score
(21.84),
followed
closely
ChatGPT
(21.83)
Gemini
(21.47).
Inappropriate
response
rates
did
not
differ
significantly,
though
trend
toward
higher
rates.
However,
accuracy
(92.5%)
responses,
(92.1%)
(90.4%).
Qualitative
highlighted
ChatGPT's
clarity
structure,
Gemini's
accessibility
but
shallowness,
Claude's
organization
occasional
divergence
focus.
LLMs
like
show
potential
supporting
require
oversight
to
ensure
performed
nearly
identically
overall,
having
score,
difference
marginal.
Further
development
is
necessary
enhance
their
for
use.
Journal of Medical Internet Research,
Journal Year:
2025,
Volume and Issue:
27, P. e63786 - e63786
Published: March 12, 2025
Effective
physician-patient
communication
is
essential
in
clinical
practice,
especially
oncology,
where
radiology
reports
play
a
crucial
role.
These
are
often
filled
with
technical
jargon,
making
them
challenging
for
patients
to
understand
and
affecting
their
engagement
decision-making.
Large
language
models,
such
as
GPT-4,
offer
novel
approach
simplifying
these
potentially
enhancing
patient
outcomes.
We
aimed
assess
the
feasibility
effectiveness
of
using
GPT-4
simplify
oncological
improve
communication.
In
retrospective
study
approved
by
ethics
review
committees
multiple
hospitals,
698
malignant
tumors
produced
between
October
2023
December
were
analyzed.
total,
70
(10%)
selected
develop
templates
scoring
scales
create
simplified
interpretative
(IRRs).
Radiologists
checked
consistency
original
IRRs,
while
volunteer
family
members
patients,
all
whom
had
at
least
junior
high
school
education
no
medical
background,
assessed
readability.
Doctors
evaluated
efficiency
through
simulated
consultations.
Transforming
into
IRRs
resulted
clearer
reports,
word
count
increasing
from
818.74
1025.82
(P<.001),
volunteers'
reading
time
decreasing
674.86
seconds
589.92
rate
72.15
words
per
minute
104.70
(P<.001).
Physician-patient
significantly
decreased,
1116.11
745.30
comprehension
scores
improved
5.51
7.83
This
demonstrates
significant
potential
large
specifically
facilitate
reports.
Simplified
enhance
understanding
doctor-patient
interactions,
suggesting
valuable
application
artificial
intelligence
practice
outcomes
health
care
