Transfer-Learning Deep Radiomics and Hand-Crafted Radiomics for Classifying Lymph Nodes from Contrast-Enhanced Computed Tomography in Lung Cancer
Cancers,
Год журнала:
2023,
Номер
15(10), С. 2850 - 2850
Опубликована: Май 21, 2023
Positron
emission
tomography
(PET)
is
currently
considered
the
non-invasive
reference
standard
for
lymph
node
(N-)staging
in
lung
cancer.
However,
not
all
patients
can
undergo
this
diagnostic
procedure
due
to
high
costs,
limited
availability,
and
additional
radiation
exposure.
The
purpose
of
study
was
predict
PET
result
from
traditional
contrast-enhanced
computed
(CT)
test
different
feature
extraction
strategies.
Язык: Английский
Prognostic Significance of 18F-FDG PET/CT Radiomics in Patients With Resectable Pancreatic Ductal Adenocarcinoma Undergoing Curative Surgery
Clinical Nuclear Medicine,
Год журнала:
2024,
Номер
49(10), С. 909 - 916
Опубликована: Июль 1, 2024
This
study
aimed
to
investigate
the
prognostic
significance
of
PET/CT
radiomics
predict
overall
survival
(OS)
in
patients
with
resectable
pancreatic
ductal
adenocarcinoma
(PDAC).
Язык: Английский
Predicting CD27 expression and clinical prognosis in serous ovarian cancer using CT-based radiomics
Journal of Ovarian Research,
Год журнала:
2024,
Номер
17(1)
Опубликована: Июнь 22, 2024
Abstract
Background
This
study
aimed
to
develop
and
evaluate
radiomics
models
predict
CD27
expression
clinical
prognosis
before
surgery
in
patients
with
serous
ovarian
cancer
(SOC).
Methods
We
used
transcriptome
sequencing
data
contrast-enhanced
computed
tomography
images
of
SOC
from
The
Cancer
Genome
Atlas
(
n
=
339)
Imaging
Archive
57)
evaluated
the
significance
prognostic
value
expression.
Radiomics
features
were
selected
create
a
recursive
feature
elimination-logistic
regression
(RFE-LR)
model
least
absolute
shrinkage
selection
operator
logistic
(LASSO-LR)
for
prediction.
Results
was
upregulated
tumor
samples,
high
level
determined
be
an
independent
protective
factor
survival.
A
set
three
six
extracted
RFE-LR
LASSO-LR
models,
respectively.
Both
demonstrated
good
calibration
benefits,
as
by
receiver
operating
characteristic
(ROC)
curves,
decision
curve
analysis.
performed
better
than
model,
owing
area
under
(AUC)
values
ROC
curves
(0.829
vs.
0.736).
Furthermore,
AUC
score
that
predicted
overall
survival
diagnosed
after
60
months
0.788
using
model.
Conclusion
we
developed
are
promising
noninvasive
tools
predicting
status
prognosis.
is
highly
recommended
evaluating
preoperative
risk
stratification
SOCs
applications.
Язык: Английский
Radiomics based on 18F-FDG PET/CT for prediction of pathological complete response to neoadjuvant therapy in non-small cell lung cancer
Frontiers in Oncology,
Год журнала:
2024,
Номер
14
Опубликована: Июль 26, 2024
Purpose
This
study
aimed
to
establish
and
evaluate
the
value
of
integrated
models
involving
18
F-FDG
PET/CT-based
radiomics
clinicopathological
information
in
prediction
pathological
complete
response
(pCR)
neoadjuvant
therapy
(NAT)
for
non-small
cell
lung
cancer
(NSCLC).
Methods
A
total
106
eligible
NSCLC
patients
were
included
study.
After
volume
interest
(VOI)
segmentation,
2,016
PET-based
CT-based
radiomic
features
extracted.
To
select
an
optimal
machine
learning
model,
a
25
constructed
based
on
five
sets
classifiers
combined
with
predictive
feature
resources,
including
alone
radiomics,
features,
features.
Area
under
curves
(AUCs)
receiver
operator
characteristic
(ROC)
used
as
main
outcome
assess
model
performance.
Results
The
hybrid
PET/CT-derived
outperformed
PET-alone
CT-alone
pCR
NAT.
Moreover,
addition
further
enhanced
performance
model.
Ultimately,
support
vector
(SVM)-based
PET/CT
presented
efficacy
AUC
0.925
(95%
CI
0.869–0.981)
training
cohort
0.863
0.740–0.985)
test
cohort.
developed
nomogram
type
was
suggested
convenient
tool
enable
clinical
application.
Conclusions
SVM
non-invasively
predict
NAC
NSCLC.
Язык: Английский
The value of 18F-fluorodeoxyglucose positron emission tomography-based radiomics in non-small cell lung cancer
Tzu Chi Medical Journal,
Год журнала:
2024,
Номер
37(1), С. 17 - 27
Опубликована: Сен. 2, 2024
Currently,
the
second
most
commonly
diagnosed
cancer
in
world
is
lung
cancer,
and
85%
of
cases
are
non-small
cell
(NSCLC).
With
growing
knowledge
oncogene
drivers
immunology,
several
novel
therapeutics
have
emerged
to
improve
prognostic
outcomes
NSCLC.
However,
treatment
remain
diverse,
an
accurate
tool
achieve
precision
medicine
unmet
need.
Radiomics,
a
method
extracting
medical
imaging
features,
promising
for
medicine.
Among
all
radiomic
tools,
18F-fluorodeoxyglucose
positron
emission
tomography
(18F-FDG
PET)-based
radiomics
provides
distinct
information
on
glycolytic
activity
heterogeneity.
In
this
review,
we
collected
relevant
literature
from
PubMed
summarized
various
applications
18F-FDG
PET-derived
improving
detection
metastasis,
subtyping
histopathologies,
characterizing
driver
mutations,
assessing
response,
evaluating
survival
Furthermore,
reviewed
values
PET-based
deep
learning.
Finally,
challenges
caveats
exist
implementation
Implementing
clinical
practice
necessary
ensure
reproducibility.
Moreover,
basic
studies
elucidating
underlying
biological
significance
lacking.
Current
inadequacies
hamper
immediate
adoption;
however,
progressively
addressing
these
issues.
remains
invaluable
indispensable
aspect
Язык: Английский