Frontiers in Oncology,
Год журнала:
2021,
Номер
11
Опубликована: Сен. 7, 2021
Purpose
This
study
aimed
to
develop
a
radiomics
nomogram
based
on
contrast-enhanced
ultrasound
(CEUS)
for
preoperatively
assessing
microvascular
invasion
(MVI)
in
hepatocellular
carcinoma
(HCC)
patients.
Methods
A
retrospective
dataset
of
313
HCC
patients
who
underwent
CEUS
between
September
20,
2016
and
March
2020
was
enrolled
our
study.
The
population
randomly
grouped
as
primary
192
validation
121
Radiomics
features
were
extracted
from
the
B-mode
(BM),
artery
phase
(AP),
portal
venous
(PVP),
delay
(DP)
images
acquired
each
patient.
After
feature
selection,
BM,
AP,
PVP,
DP
scores
(Rad-score)
constructed
dataset.
four
clinical
factors
used
multivariate
logistic
regression
analysis,
then
developed.
We
also
built
preoperative
prediction
model
comparison.
performance
evaluated
via
calibration,
discrimination,
usefulness.
Results
Multivariate
analysis
indicated
that
PVP
Rad-score,
tumor
size,
AFP
(alpha-fetoprotein)
level
independent
risk
predictors
associated
with
MVI.
incorporating
these
revealed
superior
discrimination
(based
size
level)
(AUC:
0.849
vs
.
0.690;
p
<
0.001)
0.788
0.661;
=
0.008),
good
calibration.
Decision
curve
confirmed
clinically
useful.
Furthermore,
significant
improvement
net
reclassification
index
(NRI)
integrated
discriminatory
(IDI)
implied
signatures
may
be
very
useful
biomarkers
MVI
HCC.
Conclusion
CEUS-based
showed
favorable
predictive
value
identification
could
guide
more
appropriate
surgical
planning.
The
purpose
of
this
study
was
to
develop
and
validate
a
radiomics
nomogram
for
preoperative
differentiating
focal
nodular
hyperplasia
(FNH)
from
hepatocellular
carcinoma
(HCC)
in
the
non-cirrhotic
liver.
A
total
156
patients
with
FNH
(n
=
55)
HCC
101)
were
divided
into
training
set
119)
validation
37).
Radiomics
features
extracted
triphasic
contrast
CT
images.
signature
constructed
least
absolute
shrinkage
selection
operator
algorithm,
score
(Rad-score)
calculated.
Clinical
data
findings
assessed
build
clinical
factors
model.
Combined
Rad-score
independent
factors,
by
multivariate
logistic
regression
analysis.
Nomogram
performance
respect
discrimination
usefulness.
Four
thousand
two
hundred
twenty-seven
reduced
10
as
most
important
discriminators
signature.
showed
good
(AUC
[area
under
curve],
0.964;
95%
confidence
interval
[CI],
0.934-0.995)
(AUC,
0.865;
CI,
0.725-1.000).
Age,
Hepatitis
B
virus
infection,
enhancement
pattern
factors.
nomogram,
which
incorporated
0.979;
0.959-0.998)
0.917;
0.800-1.000),
better
capability
(P
<
0.001)
compared
model
0.799;
0.719-0.879)
set.
Decision
curve
analysis
outperformed
terms
CT-based
noninvasive
prediction
tool
that
incorporates
shows
favorable
predictive
efficacy
liver,
might
facilitate
decision-making
process.
Frontiers in Oncology,
Год журнала:
2020,
Номер
10
Опубликована: Март 19, 2020
Objectives:
To
establish
a
radiomic
algorithm
based
on
grayscale
ultrasound
images
and
to
make
preoperative
predictions
of
microvascular
invasion
(MVI)
in
hepatocellular
carcinoma
(HCC)
patients.
Methods:
In
this
retrospective
study,
322
cases
histopathologically
confirmed
HCC
lesions
were
included.
The
classifications
performed
two
stages:
(1)
classifier
#1,
MVI-negative
MVI-positive
cases;
(2)
#2,
further
classified
as
M1
or
M2
cases.
gross-tumoral
region
(GTR)
peri-tumoral
(PTR)
signatures
combined
generate
gross-
(GPTR)
signatures.
optimal
incorporated
with
vital
clinical
information.
Multivariable
logistic
regression
was
used
build
models.
Results:
Finally,
1595
features
extracted
from
each
lesion.
At
the
#1
stage,
GTR,
PTR,
GPTR
showed
area
under
curve
(AUC)
values
0.708
(95
%
CI,
0.603
-
0.812),
0.710
0.609
0.811),
0.726
(95%
0.625
0.827),
respectively.
Upon
incorporation
information,
AUC
0.744
0.646
0.841).
#2
GTR
signature
0.806
0.667
0.944).
Conclusions:
Our
has
potential
value
facilitate
prediction
MVI
may
be
helpful
for
discriminating
between
levels
among
Frontiers in Oncology,
Год журнала:
2020,
Номер
10
Опубликована: Сен. 24, 2020
Preoperative
identification
of
hepatocellular
carcinoma
(HCC),
combined
hepatocellular-cholangiocarcinoma
(cHCC-ICC),
and
intrahepatic
cholangiocarcinoma
(ICC)
is
essential
for
treatment
decision
making.
We
aimed
to
use
ultrasound-based
radiomics
analysis
non-invasively
distinguish
histopathological
subtypes
primary
liver
cancer
(PLC)
before
surgery.We
retrospectively
analyzed
ultrasound
images
668
PLC
patients,
comprising
531
HCC
48
cHCC-ICC
89
ICC
patients.
The
boundary
a
tumor
was
manually
determined
on
the
largest
imaging
slice
medicine
image
by
ITK-SNAP
software
(version
3.8.0),
then,
high-throughput
features
were
extracted
from
obtained
region
interest
(ROI)
tumor.
combination
different
dimension-reduction
technologies
machine
learning
approaches
used
identify
important
develop
moderate
model.
comprehensive
ability
model
can
be
evaluated
area
under
receiver
operating
characteristic
curve
(AUC).After
digitally
processing
images,
5,234
obtained.
Spearman
+
least
absolute
shrinkage
selection
operator
(LASSO)
regression
method
feature
logistics
modeling
HCC-vs-non-HCC
(composed
16
features).
statistical
test
random
forest
methods
selection,
applied
ICC-vs-cHCC-ICC
19
overall
performance
in
identifying
types
moderate,
with
AUC
values
0.854
(training
cohort)
0.775
(test
0.920
0.728
model.Ultrasound-based
models
help
provide
effective
clinical
making
accurate
diagnosis
PLC.
Clinical Hemorheology and Microcirculation,
Год журнала:
2021,
Номер
77(4), С. 461 - 469
Опубликована: Янв. 12, 2021
To
investigate
the
clinical
value
of
dynamic
contrast
enhanced
ultrasound
(D-CEUS)
in
predicting
microvascular
invasion
(MVI)
hepatocellular
carcinoma
(HCC).In
this
retrospective
study,
16
patients
with
surgery
and
histopathologically
proved
HCC
lesions
were
included.
Patients
classified
according
to
presence
MVI:
MVI
positive
group
(n
=
6)
negative
10).
Contrast
(CEUS)
examinations
performed
within
a
week
before
surgery.
Dynamic
analysis
was
by
VueBox®
software
(Bracco,
Italy).
Three
regions
interests
(ROIs)
set
center
lesions,
at
margin
surrounding
liver
parenchyma
accordingly.
Time
intensity
curves
(TICs)
generated
quantitative
perfusion
parameters
including
WiR
(wash-in
rate),
WoR
(wash-out
WiAUC
area
under
curve),
WoAUC
curve)
WiPi
index)
obtained
analyzed.All
showed
arterial
hyperenhancement
(100
%)
late
phase
as
hypoenhancement
(75%)
CEUS.
Among
all
CEUS
parameters,
higher
than
(P
<
0.05),
WiAUC,
WiPI
lesions.
significant
group.D-CEUS
has
potential
existence
World Journal of Gastroenterology,
Год журнала:
2022,
Номер
28(20), С. 2176 - 2183
Опубликована: Май 27, 2022
Hepatocellular
carcinoma
(HCC)
is
the
most
common
primary
liver
cancer,
accounting
for
about
90%
of
cancer
cases.
It
currently
fifth
in
world
and
third
leading
cause
cancer-related
mortality.
Moreover,
recurrence
HCC
common.
Microvascular
invasion
(MVI)
a
major
factor
associated
with
postoperative
HCC.
difficult
to
evaluate
MVI
using
traditional
imaging
modalities.
Currently,
assessed
primarily
through
pathological
immunohistochemical
analyses
tissue
samples.
Needle
biopsy
method
used
confirm
diagnosis
before
surgery.
As
puncture
specimens
represent
just
small
part
tumor,
given
heterogeneity
HCC,
samples
may
yield
false-negative
results.
Radiomics,
an
emerging,
powerful,
non-invasive
tool
based
on
various
modalities,
such
as
computed
tomography,
magnetic
resonance
imaging,
ultrasound,
positron
emission
can
predict
HCC-MVI
status
preoperatively
by
delineating
tumor
and/or
regions
at
certain
distance
from
surface
extract
image
features.
Although
positive
results
have
been
reported
radiomics,
its
drawbacks
limited
clinical
translation.
This
article
reviews
application
preoperative
evaluation
explores
future
research
directions
that
facilitate
