Journal of Magnetic Resonance Imaging,
Journal Year:
2023,
Volume and Issue:
59(2), P. 613 - 625
Published: May 18, 2023
Background
Radiomics
has
been
applied
for
assessing
lymphovascular
invasion
(LVI)
in
patients
with
breast
cancer.
However,
associations
between
features
from
peritumoral
regions
and
the
LVI
status
were
not
investigated.
Purpose
To
investigate
value
of
intra‐
radiomics
LVI,
to
develop
a
nomogram
assist
making
treatment
decisions.
Study
Type
Retrospective.
Population
Three
hundred
sixteen
enrolled
two
centers
divided
into
training
(
N
=
165),
internal
validation
83),
external
68)
cohorts.
Field
Strength/Sequence
1.5
T
3.0
T/dynamic
contrast‐enhanced
(DCE)
diffusion‐weighted
imaging
(DWI).
Assessment
extracted
selected
based
on
magnetic
resonance
(MRI)
sequences
create
multiparametric
MRI
combined
signature
(RS‐DCE
plus
DWI).
The
clinical
model
was
built
MRI‐axillary
lymph
nodes
(MRI
ALN),
MRI‐reported
edema
(MPE),
apparent
diffusion
coefficient
(ADC).
constructed
RS‐DCE
DWI,
ALN,
MPE,
ADC.
Statistical
Tests
Intra‐
interclass
correlation
analysis,
Mann–Whitney
U
test,
least
absolute
shrinkage
selection
operator
regression
used
feature
selection.
Receiver
operating
characteristic
decision
curve
analyses
compare
performance
model,
nomogram.
Results
A
total
10
found
be
associated
3
7
areas.
showed
good
(AUCs,
vs.
0.884
0.695
0.870),
0.813
0.794),
0.862
0.601
0.849)
Data
Conclusion
preoperative
might
effectively
assess
LVI.
Level
Evidence
Technical
Efficacy
Stage
2
Insights into Imaging,
Journal Year:
2024,
Volume and Issue:
15(1)
Published: May 31, 2024
Abstract
Objectives
To
compare
the
diagnostic
performance
of
intratumoral
and
peritumoral
features
from
different
contrast
phases
breast
dynamic
contrast-enhanced
magnetic
resonance
imaging
(DCE-MRI)
by
building
radiomics
models
for
differentiating
molecular
subtypes
cancer.
Methods
This
retrospective
study
included
377
patients
with
pathologically
confirmed
Patients
were
divided
into
training
set
(
n
=
202),
validation
87)
test
88).
The
volume
interest
(VOI)
VOI
delineated
on
primary
cancers
at
three
DCE-MRI
phases:
early,
peak,
delayed.
Radiomics
extracted
each
phase.
After
feature
standardization,
was
filtered
variance
analysis,
correlation
least
absolute
shrinkage
selection
(LASSO).
Using
features,
a
logistic
regression
model
based
tumor
subtype
(Luminal
A,
Luminal
B,
HER2-enriched,
triple-negative)
established.
Ten
or/plus
developed
differentiation.
Results
delayed
phase
demonstrated
dominant
over
other
phases.
However,
differences
not
statistically
significant.
In
full
fusion
subtypes,
most
frequently
screened
those
According
to
Shapley
additive
explanation
(SHAP)
method,
important
also
identified
Conclusions
can
provide
additional
information
preoperative
typing.
cannot
be
ignored.
Critical
relevance
statement
constructed
played
crucial
role
in
classification,
although
no
significant
difference
observed
cohort.
Key
Points
cancer
provides
basis
setting
treatment
strategy
prognosis.
delayed-phase
outperformed
that
early-/peak-phases,
but
differently
than
or
combinations.
Both
intra-
offer
valuable
insights
Graphical
Frontiers in Cell and Developmental Biology,
Journal Year:
2024,
Volume and Issue:
11
Published: Jan. 8, 2024
This
study
explores
the
potential
of
radiomics
to
predict
proliferation
marker
protein
Ki-67
levels
and
human
epidermal
growth
factor
receptor
2
(HER-2)
status
based
on
MRI
images
patients
with
spinal
metastasis
from
primary
breast
cancer.
A
total
110
pathologically
confirmed
metastases
cancer
were
enrolled
between
Dec.
2017
2021.
All
underwent
T1-weighted
contrast-enhanced
scans.
The
PyRadiomics
package
was
used
extract
features
intraclass
correlation
coefficient
least
absolute
shrinkage
selection
operator.
most
predictive
develop
signature.
Chi-Square
test,
Fisher’s
exact
Student’s
t
-test,
Mann–Whitney
U
test
evaluate
clinical
pathological
characteristics
high-
low-level
groups
HER-2
positive/negative
groups.
models
compared
using
receiver
operating
characteristic
curve
analysis.
area
under
(AUC),
sensitivity
(SEN),
specificity
(SPE)
generated
as
comparison
metrics.
From
scans,
five
two
identified
for
level
status,
respectively.
developed
signatures
good
prediction
performance
in
training
(AUC
=
0.812,
95%
CI:
0.710–0.914,
SEN
0.667,
SPE
0.846)
validation
0.799,
0.652–0.947,
0.722,
0.833)
cohorts.
Good
also
achieved
0.796,
0.686–0.906,
0.720,
0.776)
0.705,
0.506–0.904,
0.733,
0.762)
results
this
provide
a
better
understanding
implications
MRI-based
Quantitative Imaging in Medicine and Surgery,
Journal Year:
2023,
Volume and Issue:
13(10), P. 6899 - 6910
Published: Sept. 25, 2023
The
differences
in
benign
and
malignant
breast
tumors
are
not
only
within
the
nodules
but
also
involve
changes
surrounding
tissues.
Radiomics
can
reveal
many
details
that
discernible
to
naked
eye.
This
study
aimed
distinguish
between
using
an
ultrasound-based
intra-
peritumoral
radiomics
model.This
retrospectively
collected
information
from
379
patients
with
Breast
Imaging
Reporting
Data
System
(BI-RADS)
category
3-5
clear
pathological
diagnosis
of
screened
by
routine
ultrasound
examination
Sixth
People's
Hospital
Affiliated
Medical
College
Shanghai
Jiao
Tong
University
January
2017
December
2022.
largest
dimension
lesion
on
2D
image
was
selected
outline
area
interest
which
conformally
outwardly
expanded
automatically
5
mm
extract
peritumor
features.
included
cases
were
randomly
divided
into
training
sets
test
a
ratio
7:3.
optimal
features
models
retained
statistical
machine
learning
methods
dimensionality
reduction,
logistic
regression
used
as
classifier
build
intratumoral
model
combined
intratumoral-peritumoral
model,
respectively;
through
single-factor
multifactor
regression,
could
predict
screened.
clinical
imaging
established
selecting
independent
risk
factors
univariate
multifactorial
regression.Among
BI-RADS
nodules,
there
124
255
nodules;
aged
14
88
(46.22±15.51)
years,
age
differences,
score,
mass
diameter
statistically
significant
(P>0.05).
had
under
curve
(AUC)
0.840
[95%
confidence
interval
(CI):
0.766-0.914]
set.
AUC
value
0.960
(95%
CI:
0.920-0.999).The
nomogram,
developed
features,
demonstrated
superior
performance
distinguishing
lesions.
Frontiers in Oncology,
Journal Year:
2024,
Volume and Issue:
14
Published: June 27, 2024
To
investigate
the
value
of
intralesional
and
perilesional
radiomics
based
on
computed
tomography
(CT)
in
predicting
bioactivity
hepatic
alveolar
echinococcosis
(HAE).
In
this
retrospective
study,
131
patients
who
underwent
surgical
resection
diagnosed
HAE
pathology
were
included
(bioactive,
n=69;
bioinactive,
n=62).
All
randomly
assigned
to
training
cohort
(n=78)
validation
(n=53)
a
6:4
ratio.
The
gross
lesion
volume
(GLV),
(PLV),
combined
(GPLV)
features
extracted
CT
images
portal
vein
phase.
Feature
selection
was
performed
by
intra-class
correlation
coefficient
(ICC),
univariate
analysis,
least
absolute
shrinkage
operator
(LASSO).
Radiomics
models
established
support
vector
machine
(SVM).
Radscore
best
model
clinical
independent
predictors
establish
nomogram.
Receiver
operating
characteristic
curve
(ROC)
decision
curves
used
evaluate
predictive
performance
nomogram
model.
cohort,
area
under
ROC
(AUC)
GLV,
PLV,
GPLV
radiomic
0.774,
0.729,
0.868,
respectively.
among
three
cohort.
Calcification
type
fibrinogen
(p<0.05).
AUC
nomogram-model-based
signatures
0.914
0.833
analysis
showed
that
had
greater
benefits
compared
with
single
or
shows
prediction
HAE.
including
tissue
can
significantly
improve
efficacy
bioactivity.
Medical Physics,
Journal Year:
2022,
Volume and Issue:
49(10), P. 6505 - 6516
Published: June 27, 2022
Endometrial
carcinoma
(EC)
is
one
of
the
most
common
gynecological
malignancies
with
an
increasing
incidence,
and
accurate
preoperative
diagnosis
deep
myometrial
invasion
(DMI)
crucial
for
personalized
treatment.To
determine
predictive
value
a
magnetic
resonance
imaging
(MRI)-based
radiomics
nomogram
presence
DMI
in
International
Federation
Gynecology
Obstetrics
(FIGO)
stage
I
EC.We
retrospectively
collected
163
patients
pathologically
confirmed
EC
from
two
centers
divided
all
samples
into
training
group
(Center
1)
validation
2).
Clinical
routine
indicators
were
analyzed
by
logistical
regression
to
construct
conventional
diagnostic
model
(M1).
Radiomics
features
extracted
axial
T2-weighted
contrast-enhanced
T1-weighted
(CE-T1W)
images
treated
intraclass
correlation
coefficient,
Mann-Whitney
U
test,
least
absolute
shrinkage
selection
operator,
logistic
analysis
Akaike
information
criterion
build
combined
signature
(M2).
A
(M3)
was
constructed
M1
M2.
Calibration
decision
curves
drawn
evaluate
cohorts.
The
performance
each
indicator
evaluated
area
under
receiver
operating
characteristic
curve
(AUC).The
four
significant
finally
selected
CE-T1W
MRI.
For
DMI,
AUCT
/AUCV
0.798/0.738,
M2
0.880/0.852,
M3
0.936/0.871
groups,
respectively.
calibration
showed
that
good
agreement
ideal
values.
suggested
potential
clinical
application
values
nomogram.A
based
on
MRI
can
improve
FIGO
EC.
Thoracic Cancer,
Journal Year:
2022,
Volume and Issue:
13(22), P. 3183 - 3191
Published: Oct. 6, 2022
Abstract
Background
To
evaluate
the
performances
of
multiparametric
MRI‐based
convolutional
neural
networks
(CNNs)
for
preoperative
assessment
breast
cancer
molecular
subtypes.
Methods
A
total
136
patients
with
pathologically
confirmed
invasive
cancers
were
randomly
divided
into
training,
validation,
and
testing
sets
in
this
retrospective
study.
The
CNN
models
established
based
on
contrast‐enhanced
T
1
‐weighted
imaging
(T
C),
Apparent
diffusion
coefficient
(ADC),
2
W)
using
training
validation
sets.
evaluated
set.
area
under
receiver
operating
characteristic
curve
(AUC),
sensitivity,
specificity,
accuracy
calculated
to
assess
performance.
Results
For
separation
each
subtype
from
other
subtypes
set,
C‐based
yielded
AUCs
0.762
0.920;
ADC‐based
0.686
0.851;
W‐based
achieved
0.639
0.697.
Conclusion
performed
better
than
assessing
discriminating
our
triple
negative
human
epidermal
growth
factor
receptor
2‐enriched
that
luminal
B
