arXiv (Cornell University),
Journal Year:
2022,
Volume and Issue:
unknown
Published: Jan. 1, 2022
Assessing
the
critical
view
of
safety
in
laparoscopic
cholecystectomy
requires
accurate
identification
and
localization
key
anatomical
structures,
reasoning
about
their
geometric
relationships
to
one
another,
determining
quality
exposure.
Prior
works
have
approached
this
task
by
including
semantic
segmentation
as
an
intermediate
step,
using
predicted
masks
then
predict
CVS.
While
these
methods
are
effective,
they
rely
on
extremely
expensive
ground-truth
annotations
tend
fail
when
is
incorrect,
limiting
generalization.
In
work,
we
propose
a
method
for
CVS
prediction
wherein
first
represent
surgical
image
disentangled
latent
scene
graph,
process
representation
graph
neural
network.
Our
representations
explicitly
encode
information
-
object
location,
class
information,
relations
improve
anatomy-driven
reasoning,
well
visual
features
retain
differentiability
thereby
provide
robustness
errors.
Finally,
address
annotation
cost,
train
our
only
bounding
box
annotations,
incorporating
auxiliary
reconstruction
objective
learn
fine-grained
boundaries.
We
show
that
not
outperforms
several
baseline
trained
with
but
also
scales
effectively
masks,
maintaining
state-of-the-art
performance.
Deleted Journal,
Journal Year:
2024,
Volume and Issue:
95(6), P. 429 - 435
Published: March 5, 2024
At
the
central
workplace
of
surgeon
digitalization
operating
room
has
particular
consequences
for
surgical
work.
Starting
with
intraoperative
cross-sectional
imaging
and
sonography,
through
functional
imaging,
minimally
invasive
robot-assisted
surgery
up
to
digital
anesthesiological
documentation,
vast
majority
rooms
are
now
at
least
partially
digitalized.
The
increasing
whole
process
chain
enables
not
only
collection
but
also
analysis
big
data.
Current
research
focuses
on
artificial
intelligence
data
as
prerequisite
assistance
systems
that
support
decision
making
or
warn
risks;
however,
these
technologies
raise
new
ethical
questions
community
affect
core
Artificial Intelligence in Gastrointestinal Endoscopy,
Journal Year:
2024,
Volume and Issue:
5(2)
Published: May 11, 2024
The
incidence
of
gastrointestinal
malignancies
has
increased
over
the
past
decade
at
an
alarming
rate.
Colorectal
and
gastric
cancers
are
third
fifth
most
commonly
diagnosed
worldwide
but
cited
as
second
leading
causes
mortality.
Early
institution
appropriate
therapy
from
timely
diagnosis
can
optimize
patient
outcomes.
Artificial
intelligence
(AI)-assisted
diagnostic,
prognostic,
therapeutic
tools
assist
in
expeditious
diagnosis,
treatment
planning/response
prediction,
post-surgical
prognostication.
AI
intercept
neoplastic
lesions
their
primordial
stages,
accurately
flag
suspicious
and/or
inconspicuous
with
greater
accuracy
on
radiologic,
histopathological,
endoscopic
analyses,
eliminate
over-dependence
clinicians.
AI-based
models
have
shown
to
be
par,
sometimes
even
outperformed
experienced
gastroenterologists
radiologists.
Convolutional
neural
networks
(state-of-the-art
deep
learning
models)
powerful
computational
models,
invaluable
field
precision
oncology.
These
not
only
reliably
classify
images,
also
predict
response
chemotherapy,
tumor
recurrence,
metastasis,
survival
rates
post-treatment.
In
this
systematic
review,
we
analyze
available
evidence
about
utility
artificial
PLoS ONE,
Journal Year:
2024,
Volume and Issue:
19(6), P. e0304771 - e0304771
Published: June 17, 2024
Organ
segmentation
has
become
a
preliminary
task
for
computer-aided
intervention,
diagnosis,
radiation
therapy,
and
critical
robotic
surgery.
Automatic
organ
from
medical
images
is
challenging
due
to
the
inconsistent
shape
size
of
different
organs.
Besides
this,
low
contrast
at
edges
organs
similar
types
tissue
confuses
network’s
ability
segment
contour
properly.
In
this
paper,
we
propose
novel
convolution
neural
network
based
uncertainty-driven
boundary-refined
(UDBRNet)
that
segments
CT
images.
The
are
segmented
first
produce
multiple
masks
multi-line
decoder.
Uncertain
regions
identified
boundaries
refined
on
uncertainty
data.
Our
method
achieves
remarkable
performance,
boasting
dice
accuracies
0.80,
0.95,
0.92,
0.94
Esophagus,
Heart,
Trachea,
Aorta
respectively
SegThor
dataset,
0.71,
0.89,
0.85,
0.97,
0.97
Spinal
Cord,
Left-Lung,
Right-Lung
LCTSC
dataset.
These
results
demonstrate
superiority
our
boundary
refinement
technique
over
state-of-the-art
networks
such
as
UNet,
Attention
FC-denseNet,
BASNet,
UNet++,
R2UNet,
TransUNet,
DS-TransUNet.
UDBRNet
presents
promising
more
precise
segmentation,
particularly
in
challenging,
uncertain
conditions.
source
code
proposed
will
be
available
https://github.com/riadhassan/UDBRNet
.
Artificial Intelligence Surgery,
Journal Year:
2024,
Volume and Issue:
4(4), P. 348 - 63
Published: Oct. 31, 2024
AI
is
revolutionizing
the
landscape
of
colorectal
cancer
(CRC)
surgery,
permeating
diverse
facets
ranging
from
intraoperative
guidance
to
predictive
modeling
postoperative
outcomes.
This
scoping
review
aims
comprehensively
delineate
breadth
artificial
intelligence
(AI)
applications
in
CRC
surgery.
A
search
PubMed,
Embase,
and
Ebsco
databases
up
December
2023
was
conducted,
with
registration
international
prospective
register
systematic
reviews
(PROSPERO)
(CRD42024502107).
Sixty-two
studies
meeting
stringent
inclusion
criteria
were
scrutinized,
encompassing
utilization
surgery
or
development
AI-driven
tools
for
surgical
practice.
Five
principal
domains
application
emerged:
(i)
Intraoperative
guidance,
leveraging
real-time
navigation,
indocyanine
green
(ICG)
angiography,
hyperspectral
imaging
(HSI)
enhance
precision;
(ii)
Image
segmentation,
facilitating
phase
recognition,
anatomical
identification
optimize
visualization;
(iii)
Training
performance
assessment,
enabling
objective
evaluation
enhancement
skills
through
simulations
feedback
mechanisms;
(iv)
Prediction
complications,
prognostication
anastomotic
leakage
(AL)
stricture,
stoma
requirements,
prediction
low
anterior
resection
syndrome
(LARS)
short-term
complications;
(v)
Utilization
electronic
health
records
(EHRs),
harnessing
algorithms
streamline
data
analysis
inform
decision-making
processes.
underscores
paradigm-shifting
impact
transcending
conventional
boundaries
catalyzing
advancements
across
domains.
Although
many
are
still
experimental,
as
continues
evolve,
it
promises
transform
practice,
outcomes,
revolutionize
patient
care.
Embracing
technologies
imperative
surgeons
remain
at
vanguard
innovation
deliver
superior
outcomes
patients.
Artificial Intelligence in Gastroenterology,
Journal Year:
2023,
Volume and Issue:
4(3), P. 64 - 71
Published: Dec. 7, 2023
BACKGROUND
Colorectal
cancer
is
a
major
public
health
problem,
with
1.9
million
new
cases
and
953000
deaths
worldwide
in
2020.
Total
mesorectal
excision
(TME)
the
standard
of
care
for
treatment
rectal
crucial
to
prevent
local
recurrence,
but
it
technically
challenging
surgery.
The
use
artificial
intelligence
(AI)
could
help
improve
performance
safety
TME
AIM
To
review
literature
on
AI
machine
learning
surgery
potential
future
developments.
METHODS
Online
scientific
databases
were
searched
articles
between
2020
2023.
RESULTS
search
yielded
876
results,
only
13
studies
selected
review.
specifically
rapidly
evolving
field.
There
are
number
different
algorithms
that
have
been
developed
TME,
including
instrument
detection,
anatomical
structure
identification,
image-guided
navigation
systems.
CONCLUSION
has
revolutionize
by
providing
real-time
surgical
guidance,
preventing
complications,
improving
training.
However,
further
research
needed
fully
understand
benefits
risks