Differentiation of early-stage tumors from benign lesions manifesting as pure ground-glass nodule: a clinical prediction study based on AI-derived quantitative parameters
Frontiers in Oncology,
Год журнала:
2025,
Номер
15
Опубликована: Май 19, 2025
Differentiating
between
benign
and
malignant
pure
ground-glass
nodule
(pGGN)
is
of
great
clinical
significance.
The
aim
our
study
was
to
evaluate
whether
AI-derived
quantitative
parameters
could
predict
benignity
versus
early-stage
tumors
manifesting
as
pGGN.
A
total
1,538
patients
with
pGGN
detected
by
chest
CT
at
different
campuses
hospital
from
May
2013
December
2023
were
retrospectively
analyzed.
This
included
data,
well
parameters.
All
randomly
divided
into
a
training
group
(n=893),
an
internal
validation
(n=382),
external
(n=263).
Hazard
factors
for
identified
using
univariate
analysis
multivariate
logistic
regression
analysis.
Independent
risk
then
screened,
prediction
nomogram
constructed
maximize
predictive
efficacy
application
value.
performance
the
evaluated
ROC
curves
calibration
curves,
while
decision
curve
(DCA)
used
assess
net
benefit
threshold.
final
model
nine
independent
predictors
(age,
location,
minimum
value,
standard
deviation,
kurtosis,
compactness,
energy,
costopleural
distance,
volume)
developed
user-friendly
nomogram.
AUCs
in
training,
validation,
cohorts
0.696
(95%
CI:
0.638-0.754),
0.627
0.533-0.722),
0.672
0.543-0.801),
respectively.
plot
demonstrated
good
correlation
observed
predicted
values,
remained
valid
cohort.
DCA
showed
that
model's
acceptable,
providing
substantial
application.
nomogram,
based
on
parameters,
visually
displays
overall
score
differentiate
lesions
may
serve
convenient
screening
tool
use
provides
reference
formulating
individualized
follow-up
treatment
plans
Язык: Английский
Human and Deep Learning Predictions of Peripheral Lung Cancer Using a 1.3 mm Video Endoscopic Probe
Edoardo Amante,
Robin Ghyselinck,
Luc Thiberville
и другие.
Respirology,
Год журнала:
2025,
Номер
unknown
Опубликована: Май 28, 2025
ABSTRACT
Background
and
Objective
Iriscope,
a
1.3
mm
video
endoscopic
probe
introduced
through
an
r‐EBUS
catheter,
allows
for
the
direct
visualisation
of
small
peripheral
pulmonary
nodules
(PPNs).
This
study
assessed
ability
physicians
with
different
levels
experience
in
bronchoscopy,
artificial
intelligence
(AI)
to
predict
malignant
nature
PPNs
during
Iriscope
endoscopy.
Methods
Patients
undergoing
bronchoscopy
<
20
definite
diagnosis
were
analysed.
Senior
Junior
independently
interpreted
video‐recorded
sequences,
classifying
them
as
tumoral
(malignant)
or
non‐tumoral,
blind
final
diagnosis.
A
deep
learning
(DL)
model
was
also
trained
on
images
tested
set
patients
comparison
human
interpretation.
Diagnostic
accuracy,
sensitivity,
specificity,
F1
score
calculated.
Results
Sixty‐one
(median
size
15
mm,
IQR:
11–20
mm)
included.
The
technique
allowed
lesions
all
cases.
cancer
37
cases
benign
lesion
24
outperformed
junior
recognising
images,
balanced
accuracy
85.4%
versus
66.7%,
respectively,
when
compared
DL
71.5%
but
not
superior
senior
physicians.
Conclusion
could
be
valuable
tool
management,
especially
experienced
operators.
Applied
enhance
overall
performance
less
diagnosing
malignancy.
image
Язык: Английский
The TSANZ Practical Guide for Clinicians in the Management of Screen‐ and Incidentally‐Detected Nodules
Respirology,
Год журнала:
2025,
Номер
unknown
Опубликована: Май 29, 2025
ABSTRACT
The
increasing
adoption
of
lung
cancer
screening
programs
and
advancements
in
imaging
technologies
has
significantly
increased
the
detection
pulmonary
nodules,
both
incidentally
through
screening.
This
document
provides
a
comprehensive
guide
for
clinicians
to
address
complexities
managing
indeterminate
nodules
(IPNs),
emphasising
person‐centred
multidisciplinary
care.
IPNs
are
categorised
based
on
size
morphology,
with
specific
guidelines
malignancy
risk
stratification,
diagnostic
evaluation,
follow‐up.
Dedicated
nodule
evaluation
teams
(LNETs)
meetings
(MDMs)
play
critical
role
ensuring
guideline
adherence,
streamlining
pathway,
reducing
unnecessary
investigations,
improving
outcomes.
Structured
IPN
have
demonstrated
benefits
early
detection,
improved
early‐stage
cancer,
reduced
delays
treatment
initiation.
Effective
management
strategies
include
use
standardised
reporting
templates,
utilising
validated
models
such
as
PanCan
model
agreed
protocols
follow
up
IPNs.
highlights
importance
accessing
prior
assess
growth
accounting
technical
differences
between
computed
tomography
(CT)
scans.
Any
considered
be
growing
requires
discussion
at
MDM
decision
act
tissue
biopsy
appropriate.
A
will
assist
optimising
safest
most
efficient
techniques
characteristics
profile.
By
integrating
expertise
adhering
evidence‐based
protocols,
services
can
improve
timely
diagnosis
IPNs,
minimise
over‐investigation,
reduce
chance
overdiagnosis
ultimately
enhance
patient
outcomes
survival.
Язык: Английский
External validation of the performance of commercially available deep-learning-based lung nodule detection on low-dose CT images for lung cancer screening in Japan
Japanese Journal of Radiology,
Год журнала:
2024,
Номер
unknown
Опубликована: Ноя. 30, 2024
Язык: Английский
Indolent Adenocarcinoma In Situ Mistaken as Aggressive during Lung Cancer Screening Program: To Heal or to Harm?
Surgeries,
Год журнала:
2024,
Номер
5(3), С. 867 - 874
Опубликована: Сен. 16, 2024
Background:
Low-dose
computed
tomography-based
lung
cancer
screening
(LCS)
has
demonstrated
efficacy
in
reducing
mortality.
However,
concerns
about
overdiagnosis
and
overtreatment
hinder
global
LCS
implementation.
Methods:
Here,
we
report
the
unique
case
of
a
slow-growing
1
cm
pure
ground-glass
opacity
(GGO)
lung,
known
for
15
years,
which
unexpectedly
developed
into
5
mixed
GGO
within
year,
with
an
increased
solid
component
FDG-PET
uptake.
Results:
The
patient,
asymptomatic,
underwent
right
upper
lobectomy
lymphadenectomy,
even
complicated
postoperative
chylothorax,
later
revealing
to
be
affected
by
only
unchanged
adenocarcinoma
situ
(AIS).
Conclusions:
This
serves
as
reminder
potential
behavior
pre-invasive
lesions,
can
mimic
invasive
neoplasia
may
lead
overtreatment,
underscores
challenge
distinguishing
indolent
lesions
from
potentially
aggressive
malignancies
LCS,
highlighting
need
ongoing
refinement
protocols
mitigate
this
risk.
Язык: Английский
Feasibility and potential clinical benefit of dose de-escalation in stereotactic ablative radiotherapy for lung cancer lesions with ground glass opacities
Physics and Imaging in Radiation Oncology,
Год журнала:
2024,
Номер
32, С. 100681 - 100681
Опубликована: Окт. 1, 2024
Язык: Английский