Journal of Experimental & Clinical Cancer Research,
Journal Year:
2024,
Volume and Issue:
43(1)
Published: Dec. 23, 2024
Abstract
Ductal
carcinoma
in
situ
(DCIS)
is
a
noninvasive
breast
disease
that
variably
progresses
to
invasive
cancer
(IBC).
Given
the
unpredictability
of
this
progression,
most
DCIS
patients
are
aggressively
managed
similar
IBC
patients.
Undoubtedly,
treatment
paradigm
places
many
at
risk
overtreatment
and
its
significant
consequences.
Historically,
prognostic
modeling
has
included
assessment
clinicopathological
features
genomic
markers.
Although
these
provide
valuable
insights
into
tumor
biology,
they
remain
insufficient
predict
which
will
progress
IBC.
Contemporary
work
begun
focus
on
microenvironment
surrounding
ductal
cells
for
molecular
patterns
might
progression.
In
review,
extracellular
alterations
occurring
with
malignant
transformation
from
detailed.
Not
only
do
changes
collagen
abundance,
organization,
localization
mediate
transition
IBC,
but
also
discrete
post-translational
regulation
fibers
understood
promote
invasion.
Other
matrix
proteins,
such
as
metalloproteases,
decorin,
tenascin
C,
have
been
characterized
their
role
further
demonstrate
value
matrix.
Importantly,
proteins
influence
immune
fibroblasts
toward
pro-tumorigenic
phenotypes.
Thus,
progressive
play
key
invasion
promise
development.
npj Precision Oncology,
Journal Year:
2023,
Volume and Issue:
7(1)
Published: May 29, 2023
Artificial
intelligence
methods
including
deep
neural
networks
(DNN)
can
provide
rapid
molecular
classification
of
tumors
from
routine
histology
with
accuracy
that
matches
or
exceeds
human
pathologists.
Discerning
how
make
their
predictions
remains
a
significant
challenge,
but
explainability
tools
help
insights
into
what
models
have
learned
when
corresponding
histologic
features
are
poorly
defined.
Here,
we
present
method
for
improving
DNN
using
synthetic
generated
by
conditional
generative
adversarial
network
(cGAN).
We
show
cGANs
generate
high-quality
images
be
leveraged
explaining
trained
to
classify
molecularly-subtyped
tumors,
exposing
associated
state.
Fine-tuning
through
class
and
layer
blending
illustrates
nuanced
morphologic
differences
between
tumor
subtypes.
Finally,
demonstrate
the
use
augmenting
pathologist-in-training
education,
showing
these
intuitive
visualizations
reinforce
improve
understanding
manifestations
biology.
npj Precision Oncology,
Journal Year:
2025,
Volume and Issue:
9(1)
Published: Jan. 17, 2025
Predicting
long-term
recurrence
of
disease
in
breast
cancer
(BC)
patients
remains
a
significant
challenge
for
with
early
stage
who
are
at
low
to
intermediate
risk
relapse
as
determined
using
current
clinical
tools.
Prognostic
assays
which
utilize
bulk
transcriptomics
ignore
the
spatial
context
cellular
material
and
are,
therefore,
limited
value
development
mechanistic
models.
In
this
study,
Fourier-transform
infrared
(FTIR)
chemical
images
BC
tissue
were
used
train
deep
learning
models
predict
future
recurrence.
A
number
employed,
champion
employing
two-dimensional
two-dimensional-separable
convolutional
networks
found
have
predictive
performance
ROC
AUC
approximately
0.64,
compares
well
other
clinically
prognostic
space.
All-digital
imaging
may
therefore
provide
label-free
platform
histopathological
prognosis
cancer,
opening
new
horizons
deployment
these
technologies.
Tissue Engineering Part A,
Journal Year:
2024,
Volume and Issue:
30(19-20), P. 640 - 651
Published: July 23, 2024
Oral
squamous
cell
carcinoma
(OSCC)
is
a
highly
unpredictable
disease
with
devastating
mortality
rates
that
have
not
changed
over
the
past
decades,
in
face
of
advancements
treatments
and
biomarkers,
which
improved
survival
for
other
cancers.
Delays
diagnosis
are
frequent,
leading
to
more
disfiguring
poor
outcomes
patients.
The
clinical
challenge
lies
identifying
those
patients
at
highest
risk
developing
OSCC.
epithelial
dysplasia
(OED)
precursor
OSCC
variable
behavior
across
There
no
reliable
clinical,
pathological,
histological,
or
molecular
biomarker
determine
individual
OED
Similarly,
there
robust
biomarkers
predict
treatment
This
review
aims
highlight
artificial
intelligence
(AI)-based
methods
develop
predictive
transformation
response.
Biomarkers
such
as
S100A7
demonstrate
promising
appraisal
malignant
OED.
Machine
learning-enhanced
multiplex
immunohistochemistry
workflows
examine
immune
patterns
organization
within
tumor
microenvironment
generate
outcome
predictions
immunotherapy.
Deep
learning
(DL)
an
AI-based
method
using
extended
neural
network
related
architecture
multiple
"hidden"
layers
simulated
neurons
combine
simple
visual
features
into
complex
patterns.
DL-based
digital
pathology
currently
being
developed
assess
outcomes.
integration
machine
epigenomics
epigenetic
modification
diseases
improve
our
ability
detect,
classify,
associated
marks.
Collectively,
these
tools
showcase
discovery
technology,
may
provide
potential
solution
addressing
current
limitations
predicting
behavior,
both
challenges
must
be
addressed
order
survival.
Journal of the Korean Gastric Cancer Association,
Journal Year:
2023,
Volume and Issue:
23(3), P. 410 - 410
Published: Jan. 1, 2023
Recent
advances
in
artificial
intelligence
(AI)
have
provided
novel
tools
for
rapid
and
precise
pathologic
diagnosis.
The
introduction
of
digital
pathology
has
enabled
the
acquisition
scanned
slide
images
that
are
essential
application
AI.
AI
improved
diagnosis
includes
error-free
detection
potentially
negligible
lesions,
such
as
a
minute
focus
metastatic
tumor
cells
lymph
nodes,
accurate
controversial
histologic
findings,
very
well-differentiated
carcinomas
mimicking
normal
epithelial
tissues,
pathological
subtyping
cancers.
Additionally,
utilization
algorithms
enables
decision
score
immunohistochemical
markers
targeted
therapies,
human
epidermal
growth
factor
receptor
2
programmed
death-ligand
1.
Studies
revealed
assistance
can
reduce
discordance
interpretation
between
pathologists
more
accurately
predict
clinical
outcomes.
Several
approaches
been
employed
to
develop
biomarkers
from
using
Moreover,
AI-assisted
analysis
cancer
microenvironment
showed
distribution
tumor-infiltrating
lymphocytes
was
related
response
immune
checkpoint
inhibitor
therapy,
emphasizing
its
value
biomarker.
As
numerous
studies
demonstrated
significance
biomarker
development,
AI-based
approach
will
advance
diagnostic
pathology.
Science Advances,
Journal Year:
2024,
Volume and Issue:
10(46)
Published: Nov. 15, 2024
Artificial
intelligence
models
have
been
increasingly
used
in
the
analysis
of
tumor
histology
to
perform
tasks
ranging
from
routine
classification
identification
molecular
features.
These
approaches
distill
cancer
histologic
images
into
high-level
features,
which
are
predictions,
but
understanding
biologic
meaning
such
features
remains
challenging.
We
present
and
validate
a
custom
generative
adversarial
network—HistoXGAN—capable
reconstructing
representative
using
feature
vectors
produced
by
common
extractors.
evaluate
HistoXGAN
across
29
subtypes
demonstrate
that
reconstructed
retain
information
regarding
grade,
subtype,
gene
expression
patterns.
leverage
illustrate
underlying
for
deep
learning
actionable
mutations,
identify
model
reliance
on
batch
effect
accurate
reconstruction
radiographic
imaging
“virtual
biopsy.”
