bioRxiv (Cold Spring Harbor Laboratory),
Год журнала:
2023,
Номер
unknown
Опубликована: Ноя. 29, 2023
ABSTRACT
With
the
widespread
use
of
high-throughput
sequencing
technologies,
understanding
biology
and
cancer
heterogeneity
has
been
revolutionized.
Recently,
several
machine-learning
models
based
on
transcriptional
data
have
developed
to
accurately
predict
patient’s
outcome
clinical
response.
However,
an
open-source
R
package
covering
state-of-the-art
machine
learning
algorithms
for
user-friendly
access
yet
be
developed.
Thus,
we
proposed
a
flexible
computational
framework
construct
learning-based
integration
model
with
elegant
performance
(Mime).
Mime
streamlined
process
developing
predictive
high
accuracy,
leveraging
complex
datasets
identify
critical
genes
associated
prognosis.
An
in
silico
combined
de
novo
PIEZO1-associated
signatures
constructed
by
demonstrated
accuracy
predicting
outcomes
patients
compared
other
published
models.
In
addition,
could
also
precisely
infer
immunotherapy
response
applying
different
Mime.
Finally,
SDC1
selected
from
presented
high-potential
role
glioma
targeted
prospect.
Taken
together,
our
provides
solution
constructing
will
greatly
expanded
provide
valuable
insights
into
current
fields.
Computational and Structural Biotechnology Journal,
Год журнала:
2024,
Номер
23, С. 2798 - 2810
Опубликована: Июнь 29, 2024
The
widespread
use
of
high-throughput
sequencing
technologies
has
revolutionized
the
understanding
biology
and
cancer
heterogeneity.
Recently,
several
machine-learning
models
based
on
transcriptional
data
have
been
developed
to
accurately
predict
patients'
outcome
clinical
response.
However,
an
open-source
R
package
covering
state-of-the-art
algorithms
for
user-friendly
access
yet
be
developed.
Thus,
we
proposed
a
flexible
computational
framework
construct
machine
learning-based
integration
model
with
elegant
performance
(Mime).
Mime
streamlines
process
developing
predictive
high
accuracy,
leveraging
complex
datasets
identify
critical
genes
associated
prognosis.
An
in
silico
combined
de
novo
PIEZO1-associated
signatures
constructed
by
demonstrated
accuracy
predicting
outcomes
patients
compared
other
published
models.
Furthermore,
could
also
precisely
infer
immunotherapy
response
applying
different
Mime.
Finally,
SDC1
selected
from
potential
as
glioma
target.
Taken
together,
our
provides
solution
constructing
will
greatly
expanded
provide
valuable
insights
into
current
fields.
is
available
GitHub
(https://github.com/l-magnificence/Mime).
Diagnostics,
Год журнала:
2024,
Номер
14(2), С. 152 - 152
Опубликована: Янв. 9, 2024
Purpose:
We
aimed
to
assess
the
efficacy
of
machine
learning
and
radiomics
analysis
using
magnetic
resonance
imaging
(MRI)
with
a
hepatospecific
contrast
agent,
in
pre-surgical
setting,
predict
tumor
budding
liver
metastases.
Methods:
Patients
MRI
setting
were
retrospectively
enrolled.
Manual
segmentation
was
made
by
means
3D
Slicer
image
computing,
851
features
extracted
as
median
values
PyRadiomics
Python
package.
Balancing
performed
inter-
intraclass
correlation
coefficients
calculated
between
observer
within
reproducibility
all
features.
A
Wilcoxon–Mann–Whitney
nonparametric
test
receiver
operating
characteristics
(ROC)
carried
out.
feature
selection
procedures
performed.
Linear
non-logistic
regression
models
(LRM
NLRM)
different
learning-based
classifiers
including
decision
tree
(DT),
k-nearest
neighbor
(KNN)
support
vector
(SVM)
considered.
Results:
The
internal
training
set
included
49
patients
119
validation
cohort
consisted
total
28
single
lesion
patients.
best
predictor
classify
original_glcm_Idn
obtained
T1-W
VIBE
sequence
arterial
phase
an
accuracy
84%;
wavelet_LLH_firstorder_10Percentile
portal
92%;
wavelet_HHL_glcm_MaximumProbability
hepatobiliary
excretion
88%;
wavelet_LLH_glcm_Imc1
T2-W
SPACE
sequences
88%.
Considering
linear
analysis,
statistically
significant
increase
96%
weighted
combination
13
radiomic
from
phase.
Moreover,
classifier
KNN
trained
sequence,
obtaining
95%
AUC
0.96.
reached
94%,
sensitivity
86%
specificity
95%.
Conclusions:
Machine
are
promising
tools
predicting
budding.
there
compared
feature.
Journal of Clinical Medicine,
Год журнала:
2024,
Номер
13(2), С. 547 - 547
Опубликована: Янв. 18, 2024
Background:
Small
renal
masses
(SRMs)
are
defined
as
contrast-enhanced
lesions
less
than
or
equal
to
4
cm
in
maximal
diameter,
which
can
be
compatible
with
stage
T1a
cell
carcinomas
(RCCs).
Currently,
50–61%
of
all
tumors
found
incidentally.
Methods:
The
characteristics
the
lesion
influence
choice
type
management,
include
several
methods
SRM
including
nephrectomy,
partial
ablation,
observation,
and
also
stereotactic
body
radiotherapy.
Typical
imaging
available
for
differentiating
benign
from
malignant
ultrasound
(US),
(CEUS),
computed
tomography
(CT),
magnetic
resonance
(MRI).
Results:
Although
is
first
technique
used
detect
small
lesions,
it
has
limitations.
CT
main
most
widely
characterization.
advantages
MRI
compared
better
contrast
resolution
tissue
characterization,
use
functional
sequences,
possibility
performing
examination
patients
allergic
iodine-containing
medium,
absence
exposure
ionizing
radiation.
For
a
correct
evaluation
during
follow-up,
necessary
reliable
method
assessment
represented
by
Bosniak
classification
system.
This
was
initially
developed
based
on
findings,
2019
revision
proposed
inclusion
features;
however,
latest
not
yet
received
widespread
validation.
Conclusions:
radiomics
an
emerging
increasingly
central
field
applications
such
characterizing
masses,
distinguishing
RCC
subtypes,
monitoring
response
targeted
therapeutic
agents,
prognosis
metastatic
context.
BMC Gastroenterology,
Год журнала:
2025,
Номер
25(1)
Опубликована: Март 20, 2025
Colorectal
liver
metastases
(CRLM)
are
a
major
determinant
of
prognosis
in
colorectal
cancer
(CRC)
patients.
Their
early
and
accurate
detection
is
essential
for
appropriate
therapeutic
planning
improving
survival
outcomes.
To
evaluate
the
diagnostic
capabilities
contrast-enhanced
computed
tomography
(CT)
magnetic
resonance
imaging
(MRI)
detecting
metastases.
We
employed
case-control
design
to
compare
patients
with
histologically
confirmed
against
control
group
without
condition.
A
total
85
each
were
selected
retrospectively
matched
based
on
relevant
factors.
All
subjects
underwent
both
CT
MRI.
The
performance
these
modalities
was
assessed
by
analysing
sensitivity,
specificity,
positive
negative
predictive
values,
radiologists'
confidence.
Kappa
statistics
used
inter-observer
agreement.
MRI
scans
performed
using
3-Tesla
(3-T)
scanner
ensure
high-quality
detailed
lesion
characterization.
And
all
reviewed
two
radiologists.
combination
demonstrated
statistically
significant
improvement
sensitivity
(90.6%
alone
vs.
96.5%
combined
modalities)
specificity
(95.3%
98.3%
modalities).
Positive
values
similarly
enhanced.
Radiologists'
confidence
higher
imaging,
achieving
'very
high'
level
78.8%
cases
compared
64.7%
alone.
agreement
reached
'almost
perfect'
status
approach.
integration
significantly
enhanced
accuracy
metastases,
representing
valuable
tool
preoperative
evaluation
CRC.
Not
applicable.
Sensors,
Год журнала:
2025,
Номер
25(11), С. 3415 - 3415
Опубликована: Май 29, 2025
The
number
of
US
exams
has
nearly
doubled
in
the
last
ten
years.
Many
researchers
point
out
probe
pressure
force
influence
on
image
quality
and
other
aspects
examination.
This
review
aims
to
identify
range
applied
during
examinations
gather
information
compression
values
various
(examination
types,
body
regions,
etc.).
Methods:
A
systematic
following
PRISMA
guidelines
was
conducted
using
IEEE
Xplore,
Web
Science,
Scopus,
PubMed/MEDLINE.
Studies
with
quantitative
data
by
human
operators
or
RUSs
(robotic
ultrasound
systems)
were
included.
Results:
From
26
included
studies,
ranges
varied
up
34.5
N
for
abdominal
exams.
Robotic
systems
slightly
higher
maximum
forces
(34.5
N)
than
(30
N).
Most
studies
reported
positive
impacts
monitoring
diagnostic
precision,
no
adverse
effects
patient
comfort.
Conclusions:
evidence
collectively
emphasizes
critical
role
US.
nonuniformity
reviewed
does
not
allow
identifying
a
clearly
defined
protocols.
Integrating
RUS
standardized
protocols
could
improve
consistency
accuracy.
Radiomics
represents
the
science
of
extraction
and
analysis
a
multitude
quantitative
features
from
medical
imaging,
unrevealing
potentiality
Radiologic
images.
This
scientific
review
aims
to
provide
radiologists
comprehensive
understanding
radiomics,
emphasizing
its
principles,
applications,
challenges,
limits,
future
prospects.
Limits
standardization
actual
production
are
analyzed
with
possible
solutions
proposed
by
some
papers
reported
commented.
As
continuous
evolution
images
is
ongoing,
must
be
aware
new
perspective
perform
central
role
in
patients
management.