In Vivo,
Journal Year:
2024,
Volume and Issue:
38(4), P. 1649 - 1659
Published: Jan. 1, 2024
Background/Aim:
Demographic
change
and
increasing
complexity
of
therapy
decisions
lead
to
a
growing
burden
on
the
healthcare
system,
necessitating
efforts
simplify
enhance
efficiency
patient
care.
The
present
study
evaluates
ChatGPT's
ability
provide
recommendations
for
gynecological
malignancies
that
are
both
in
line
with
local
guidelines
individually
tailored
patient.
Patients
Methods:
Sixteen
patients
endometrial,
cervical,
ovarian
cancer
who
were
treated
clinic
University
Hospital
Erlangen
from
January
2022
August
2023
included
analysis.
Data
collected
within
clinical
routine
care
communicated
chat-based
AI
model
ChatGPT
(version
3.5).
performance
generating
treatment
plans
evaluated
using
an
answer
scoring
system
descriptive
Results:
According
[range:
−1
point
(minimum)
2
points
(maximum)],
demonstrated
good
potential
average
score
0.75
cancer,
0.7
cervical
1.5
endometrial
patients.
most
common
deductions
about
incomplete
recommendations,
whereas
contraindicated
modalities
rarely
suggested.
Individual
characteristics
regularly
considered
by
ChatGPT.
reliably
indicated
aftercare
provided
detailed
information
preventive
measures
as
well
supportive
treatment.
Conclusion:
is
promising
tool
generation
suggestions
carcinomas
high
flexibility
response
individual
differences.
At
current
state,
however,
not
suitable
replacing
expert
panels.
Journal of Cancer Research and Clinical Oncology,
Journal Year:
2024,
Volume and Issue:
150(3)
Published: March 19, 2024
Despite
advanced
technologies
in
breast
cancer
management,
challenges
remain
efficiently
interpreting
vast
clinical
data
for
patient-specific
insights.
We
reviewed
the
literature
on
how
large
language
models
(LLMs)
such
as
ChatGPT
might
offer
solutions
this
field.
JMIR AI,
Journal Year:
2024,
Volume and Issue:
3, P. e55957 - e55957
Published: May 6, 2024
Clinical
decision-making
is
a
crucial
aspect
of
health
care,
involving
the
balanced
integration
scientific
evidence,
clinical
judgment,
ethical
considerations,
and
patient
involvement.
This
process
dynamic
multifaceted,
relying
on
clinicians’
knowledge,
experience,
intuitive
understanding
to
achieve
optimal
outcomes
through
informed,
evidence-based
choices.
The
advent
generative
artificial
intelligence
(AI)
presents
revolutionary
opportunity
in
decision-making.
AI’s
advanced
data
analysis
pattern
recognition
capabilities
can
significantly
enhance
diagnosis
treatment
diseases,
processing
vast
medical
identify
patterns,
tailor
treatments,
predict
disease
progression,
aid
proactive
management.
However,
incorporation
AI
into
raises
concerns
regarding
reliability
accuracy
AI-generated
insights.
To
address
these
concerns,
11
“verification
paradigms”
are
proposed
this
paper,
with
each
paradigm
being
unique
method
verify
nature
paper
also
frames
concept
“clinically
explainable,
fair,
responsible,
clinician-,
expert-,
patient-in-the-loop
AI.”
model
focuses
ensuring
comprehensibility,
collaborative
nature,
grounding,
advocating
for
serve
as
an
augmentative
tool,
its
processes
transparent
understandable
clinicians
patients.
should
enhance,
not
replace,
clinician’s
judgment
involve
continuous
learning
adaptation
based
real-world
legal
compliance.
In
conclusion,
while
holds
immense
promise
enhancing
decision-making,
it
essential
ensure
that
produces
evidence-based,
reliable,
impactful
knowledge.
Using
outlined
paradigms
approaches
help
communities
harness
potential
maintaining
high
care
standards.
Frontiers in Oncology,
Journal Year:
2024,
Volume and Issue:
14
Published: May 24, 2024
Background
Head
and
neck
squamous
cell
carcinoma
(HNSCC)
is
a
complex
malignancy
that
requires
multidisciplinary
approach
in
clinical
practice,
especially
tumor
board
discussions.
In
recent
years,
artificial
intelligence
has
emerged
as
tool
to
assist
healthcare
professionals
making
informed
decisions.
This
study
investigates
the
application
of
ChatGPT
3.5
4.0,
natural
language
processing
models,
decision-making.
Methods
We
conducted
pilot
October
2023
on
20
consecutive
head
cancer
patients
discussed
our
(MDT).
Patients
with
primary
diagnosis
were
included.
The
MDT
4.0
recommendations
for
each
patient
compared
by
two
independent
reviewers
number
therapy
options,
recommendation,
explanation
summarization
graded.
Results
this
study,
provided
mostly
general
answers
surgery,
chemotherapy,
radiation
therapy.
For
scored
well,
but
demonstrated
be
an
assisting
tool,
suggesting
significantly
more
options
than
MDT,
while
some
recommended
treatment
modalities
like
immunotherapy
are
not
part
current
guidelines.
Conclusions
research
demonstrates
advanced
AI
models
at
moment
can
merely
setting,
since
versions
list
common
sometimes
recommend
incorrect
case
lack
information
source
material.
Journal of Sports Science and Medicine,
Journal Year:
2024,
Volume and Issue:
unknown, P. 56 - 72
Published: Jan. 12, 2024
ChatGPT
may
be
used
by
runners
to
generate
training
plans
enhance
performance
or
health
aspects.
However,
the
quality
of
generated
based
on
different
input
information
is
unknown.
The
objective
study
was
evaluate
ChatGPT-generated
six-week
for
granularity.
Three
were
using
22
criteria
drawn
from
literature
and
coaching
experts
a
1-5
Likert
Scale.
A
Friedmann
test
assessed
significant
differences
in
between
plans.
For
1,
2
3,
median
rating
<3
given
19,
11,
1
times,
3
5,
8
times
>3
0,
6,
13
respectively.
Training
plan
received
significantly
lower
ratings
compared
criteria,
15
(p
<
0.05).
0.05)
9
criteria.
are
ranked
sub-optimally
experts,
although
increases
when
more
provided.
An
understanding
aspects
relevant
programming
distance
running
important,
we
advise
avoiding
use
without
an
expert
coach's
feedback.
Archives of Gynecology and Obstetrics,
Journal Year:
2024,
Volume and Issue:
310(1), P. 537 - 550
Published: May 29, 2024
Abstract
Purpose
This
study
investigated
the
concordance
of
five
different
publicly
available
Large
Language
Models
(LLM)
with
recommendations
a
multidisciplinary
tumor
board
regarding
treatment
for
complex
breast
cancer
patient
profiles.
Methods
Five
LLM,
including
three
versions
ChatGPT
(version
4
and
3.5,
data
access
until
September
3021
January
2022),
Llama2,
Bard
were
prompted
to
produce
20
LLM
compared
(gold
standard),
surgical,
endocrine
systemic
treatment,
radiotherapy,
genetic
testing
therapy
options.
Results
GPT4
demonstrated
highest
(70.6%)
invasive
profiles,
followed
by
GPT3.5
2021
(58.8%),
2022
(41.2%),
Llama2
(35.3%)
(23.5%).
Including
precancerous
lesions
ductal
carcinoma
in
situ,
identical
ranking
was
reached
lower
overall
each
(GPT4
60.0%,
50.0%,
35.0%,
30.0%,
20.0%).
achieved
full
(100%)
radiotherapy.
Lowest
alignment
recommending
testing,
demonstrating
varying
(55.0%
2022,
up
85.0%
GPT4).
Conclusion
early
feasibility
is
first
compare
care
regard
changes
accuracy
over
time,
i.e.,
more
or
through
technological
upgrades.
Methodological
advancement,
optimization
prompting
techniques,
development,
enabling
input
control
secure
processing,
are
necessary
preparation
large-scale
multicenter
studies
provide
evidence
on
their
safe
reliable
clinical
application.
At
present,
evidenced
use
not
yet
feasible.
Journal of Surgical Oncology,
Journal Year:
2024,
Volume and Issue:
130(2), P. 188 - 203
Published: June 4, 2024
Artificial
intelligence
(AI)-driven
chatbots,
capable
of
simulating
human-like
conversations,
are
becoming
more
prevalent
in
healthcare.
While
this
technology
offers
potential
benefits
patient
engagement
and
information
accessibility,
it
raises
concerns
about
misuse,
misinformation,
inaccuracies,
ethical
challenges.
Current Oncology,
Journal Year:
2024,
Volume and Issue:
31(9), P. 4984 - 5007
Published: Aug. 27, 2024
The
integration
of
multidisciplinary
tumor
boards
(MTBs)
is
fundamental
in
delivering
state-of-the-art
cancer
treatment,
facilitating
collaborative
diagnosis
and
management
by
a
diverse
team
specialists.
Despite
the
clear
benefits
personalized
patient
care
improved
outcomes,
increasing
burden
on
MTBs
due
to
rising
incidence
financial
constraints
necessitates
innovative
solutions.
advent
artificial
intelligence
(AI)
medical
field
offers
promising
avenue
support
clinical
decision-making.
This
review
explores
perspectives
clinicians
dedicated
patients-surgeons,
oncologists,
radiation
oncologists-on
application
AI
within
MTBs.
Additionally,
it
examines
role
across
various
specialties
involved
treatment.
By
analyzing
both
potential
challenges,
this
study
underscores
how
can
enhance
discussions
optimize
treatment
plans.
findings
highlight
transformative
that
may
play
refining
oncology
sustaining
efficacy
amidst
growing
demands.
Cancers,
Journal Year:
2025,
Volume and Issue:
17(2), P. 197 - 197
Published: Jan. 9, 2025
In
recent
years,
Artificial
Intelligence
(AI)
has
shown
transformative
potential
in
advancing
breast
cancer
care
globally.
This
scoping
review
seeks
to
provide
a
comprehensive
overview
of
AI
applications
care,
examining
how
they
could
reshape
diagnosis,
treatment,
and
management
on
worldwide
scale
discussing
both
the
benefits
challenges
associated
with
their
adoption.
accordance
PRISMA-ScR
ensuing
guidelines
reviews,
PubMed,
Web
Science,
Cochrane
Library,
Embase
were
systematically
searched
from
inception
end
May
2024.
Keywords
included
"Artificial
Intelligence"
"Breast
Cancer".
Original
studies
based
focus
narrative
synthesis
was
employed
for
data
extraction
interpretation,
findings
organized
into
coherent
themes.
Finally,
84
articles
included.
The
majority
conducted
developed
countries
(n
=
54).
publications
last
10
years
83).
six
main
themes
screening
32),
image
detection
nodal
status
7),
AI-assisted
histopathology
8),
assessing
post-neoadjuvant
chemotherapy
(NACT)
response
23),
margin
assessment
5),
as
clinical
decision
support
tool
9).
been
used
tools
augment
treatment
decisions
multidisciplinary
tumor
board
settings.
Overall,
demonstrated
improved
accuracy
efficiency;
however,
most
did
not
report
patient-centric
outcomes.
show
promise
enhancing
diagnostic
planning.
However,
persistent
adoption,
such
quality,
algorithm
transparency,
resource
disparities,
must
be
addressed
advance
field.
Journal of Clinical Medicine,
Journal Year:
2025,
Volume and Issue:
14(2), P. 399 - 399
Published: Jan. 10, 2025
Background/Objectives:
The
aim
of
this
study
was
to
analyze
whether
the
implementation
artificial
intelligence
(AI),
specifically
Natural
Language
Processing
(NLP)
branch
developed
by
OpenAI,
could
help
a
thoracic
multidisciplinary
tumor
board
(MTB)
make
decisions
if
provided
with
all
patient
data
presented
committee
and
supported
accepted
clinical
practice
guidelines.
Methods:
This
is
retrospective
comparative
study.
inclusion
criteria
were
defined
as
patients
who
at
MTB
suspicious
or
first
diagnosis
non-small-cell
lung
cancer
between
January
2023
June
2023.
Intervention:
GPT
3.5
turbo
chat
used,
providing
case
summary
in
proceedings
latest
SEPAR
treatment
application
asked
issue
one
following
recommendations:
follow-up,
surgery,
chemotherapy,
radiotherapy,
chemoradiotherapy.
Statistical
analysis:
A
concordance
analysis
performed
measuring
Kappa
coefficient
evaluate
consistency
results
AI
committee's
decision.
Results:
Fifty-two
included
had
an
overall
76%,
index
0.59
replicability
92.3%
for
whom
it
recommended
surgery
(after
repeating
cases
four
times).
Conclusions:
interesting
tool
which
decision
making
MTBs.
