Radiation Oncology,
Journal Year:
2024,
Volume and Issue:
19(1)
Published: Dec. 21, 2024
To
investigate
the
early
predictive
value
of
dynamic
magnetic
resonance
imaging
(MRI)-based
radiomics
for
progression
and
prognosis
in
locally
advanced
cervical
cancer
(LACC)
patients
treated
with
concurrent
chemoradiotherapy
(CCRT).
A
total
111
LACC
(training
set:
88;
test
23)
were
retrospectively
enrolled.
Dynamic
MR
images
acquired
at
baseline
(MRIpre),
before
brachytherapy
delivery
(MRImid)
each
follow-up
visit.
Clinical
characteristics,
2-year
progression-free
survival
(PFS),
overall
(OS)
evaluated.
The
least
absolute
shrinkage
selection
operator
(LASSO)
method
was
applied
to
extract
features
from
as
well
clinical
characteristics.
support
vector
machine
(SVM)
model
trained
on
training
set
then
evaluated
set.
Compared
single-sequence
models,
multisequence
models
exhibited
superior
performance.
MRImid-based
performed
better
predicting
than
post-treatment
did.
MRIpre-,
MRImid-
ΔMRImid
(variations
MRIpre
MRImid)
-based
achieve
AUC
scores
0.723,
0.750
0.759
PFS
0.711,
0.737
0.789
OS
When
combined
ΔMRImid-based
also
other
did,
an
0.812
0.868
survival.
We
built
learning
longitudinal
found
that
can
serve
a
non-invasive
indicator
prediction
receiving
CCRT.
integrated
characteristics
further
enhanced
