Journal of Medical Imaging,
Journal Year:
2023,
Volume and Issue:
10(05)
Published: July 31, 2023
Artificial
intelligence
(AI)
presents
an
opportunity
in
anatomic
pathology
to
provide
quantitative
objective
support
a
traditionally
subjective
discipline,
thereby
enhancing
clinical
workflows
and
enriching
diagnostic
capabilities.
AI
requires
access
digitized
materials,
which,
at
present,
are
most
commonly
generated
from
the
glass
slide
using
whole-slide
imaging.
Models
developed
collaboratively
or
sourced
externally,
best
practices
suggest
validation
with
internal
datasets
closely
resembling
data
expected
practice.
Although
array
of
models
that
operational
for
improve
quality
capabilities
has
been
described,
them
can
be
categorized
into
one
more
discrete
types.
However,
their
function
workflow
vary,
as
single
algorithm
may
appropriate
screening
triage,
assistance,
virtual
second
opinion,
other
uses
depending
on
how
it
is
implemented
validated.
Despite
promise
AI,
barriers
adoption
have
numerous,
which
inclusion
new
stakeholders
expansion
reimbursement
opportunities
among
impactful
solutions.
Cell,
Journal Year:
2023,
Volume and Issue:
186(8), P. 1729 - 1754
Published: April 1, 2023
Pancreatic
ductal
adenocarcinoma
(PDAC)
remains
one
of
the
deadliest
cancers.
Significant
efforts
have
largely
defined
major
genetic
factors
driving
PDAC
pathogenesis
and
progression.
tumors
are
characterized
by
a
complex
microenvironment
that
orchestrates
metabolic
alterations
supports
milieu
interactions
among
various
cell
types
within
this
niche.
In
review,
we
highlight
foundational
studies
driven
our
understanding
these
processes.
We
further
discuss
recent
technological
advances
continue
to
expand
complexity.
posit
clinical
translation
research
endeavors
will
enhance
currently
dismal
survival
rate
recalcitrant
disease.
Nature Communications,
Journal Year:
2023,
Volume and Issue:
14(1)
Published: Dec. 8, 2023
Spatial
transcriptomics
(ST)
technologies
enable
high
throughput
gene
expression
characterization
within
thin
tissue
sections.
However,
comparing
spatial
observations
across
sections,
samples,
and
remains
challenging.
To
address
this
challenge,
we
develop
STalign
to
align
ST
datasets
in
a
manner
that
accounts
for
partially
matched
sections
other
local
non-linear
distortions
using
diffeomorphic
metric
mapping.
We
apply
as
well
3D
common
coordinate
framework.
show
achieves
cell-type
correspondence
locations
is
significantly
improved
over
landmark-based
affine
alignments.
Applying
of
the
mouse
brain
framework
from
Allen
Brain
Atlas,
highlight
how
can
be
used
lift
region
annotations
interrogation
compositional
heterogeneity
anatomical
structures.
available
an
open-source
Python
toolkit
at
https://github.com/JEFworks-Lab/STalign
Supplementary
Software
with
additional
documentation
tutorials
https://jef.works/STalign
.
Cell,
Journal Year:
2024,
Volume and Issue:
187(10), P. 2502 - 2520.e17
Published: May 1, 2024
Human
tissue,
which
is
inherently
three-dimensional
(3D),
traditionally
examined
through
standard-of-care
histopathology
as
limited
two-dimensional
(2D)
cross-sections
that
can
insufficiently
represent
the
tissue
due
to
sampling
bias.
To
holistically
characterize
histomorphology,
3D
imaging
modalities
have
been
developed,
but
clinical
translation
hampered
by
complex
manual
evaluation
and
lack
of
computational
platforms
distill
insights
from
large,
high-resolution
datasets.
We
present
TriPath,
a
deep-learning
platform
for
processing
volumes
efficiently
predicting
outcomes
based
on
morphological
features.
Recurrence
risk-stratification
models
were
trained
prostate
cancer
specimens
imaged
with
open-top
light-sheet
microscopy
or
microcomputed
tomography.
By
comprehensively
capturing
morphologies,
volume-based
prognostication
achieves
superior
performance
traditional
2D
slice-based
approaches,
including
clinical/histopathological
baselines
six
certified
genitourinary
pathologists.
Incorporating
greater
volume
improves
prognostic
mitigates
risk
prediction
variability
bias,
further
emphasizing
value
larger
extents
heterogeneous
morphology.
