CT-based radiomics models using intralesional and different perilesional signatures in predicting the microvascular density of hepatic alveolar echinococcosis
Juan Hou,
Simiao Zhang,
Shouxian Li
и другие.
BMC Medical Imaging,
Год журнала:
2025,
Номер
25(1)
Опубликована: Март 10, 2025
To
evaluate
the
performance
of
CT-based
intralesional
combined
with
different
perilesional
radiomics
models
in
predicting
microvascular
density
(MVD)
hepatic
alveolar
echinococcosis
(HAE).
This
study
retrospectively
analyzed
preoperative
CT
data
from
303
patients
HAE
confirmed
by
surgical
pathology
(MVD
positive,
n
=
182;
MVD
negative,
121).
The
were
randomly
divided
into
training
cohort
(n
242)
and
test
61)
at
a
ratio
8:2.
features
extracted
images
on
portal
vein
phase.
Four
constructed
based
gross
lesion
volume
(GLV),
10
mm
(GPLV10mm),
15
(GPLV15mm)
20
(GPLV20mm).
best
signature
model
clinical
to
establish
nomogram.
Receiver
operating
characteristic
curve
(ROC)
decision
analysis
(DCA)
used
predictive
models.
Among
four
models,
GPLV20mm
performed
highest
prediction
area
under
curves
(AUCs)
was
0.876
0.802,
respectively.
AUC
0.753
0.699
cohort.
nomogram
radiomic
signatures
0.922
0.849
DCA
showed
that
had
greater
benefits
among
three
can
better
predict
HAE.
performance.
Язык: Английский
Research Progress on the Application of Radiomics in the Diagnosis and Treatment of Gliomas
Journal of Clinical Personalized Medicine,
Год журнала:
2025,
Номер
04(02), С. 657 - 664
Опубликована: Янв. 1, 2025
Язык: Английский
Head and Neck Squamous Cell Carcinoma: Insights from Dual-Energy Computed Tomography (DECT)
Tomography,
Год журнала:
2024,
Номер
10(11), С. 1780 - 1797
Опубликована: Ноя. 11, 2024
Head
and
neck
cancer
represents
the
seventh
most
common
neoplasm
worldwide,
with
squamous
cell
carcinoma
being
represented
histologic
variant.
The
rising
incidence
of
neoplastic
pathology
this
district,
coupled
drastic
changes
in
its
epidemiology
over
past
decades,
have
posed
significant
challenges
to
physicians
worldwide
terms
diagnosis,
prognosis,
treatment.
In
order
meet
these
challenges,
a
considerable
amount
effort
has
been
spent
by
authors
recent
literature
explore
new
technologies
their
possible
employment
for
better
diagnostic
prognostic
definition
head
(HNSCC).
Among
technologies,
growing
interest
gathering
around
applications
dual-energy
computed
tomography
(DECT)
pathology.
Dual-energy
utilizes
two
distinct
X-ray
energy
spectra
obtain
datasets
single
scan,
allowing
material
differentiation
based
on
unique
attenuation
profiles.
DECT
offers
key
benefits
such
as
enhanced
contrast
resolution,
reduced
beam-hardening
artifacts,
precise
iodine
quantification
through
monochromatic
reconstructions.
It
also
creates
decomposition
images,
like
maps,
aiding
tumor
characterization
therapy
assessment.
This
paper
aims
summarize
findings
use
HNSCC,
providing
comprehensive
overview
aid
further
research
exploration
field.
Язык: Английский