Cancer and Metastasis Reviews,
Год журнала:
2025,
Номер
44(1)
Опубликована: Янв. 31, 2025
Abstract
CT
chest
scans
are
commonly
performed
worldwide,
either
in
routine
clinical
practice
for
a
wide
range
of
indications
or
as
part
lung
cancer
screening
programs.
Many
these
detect
nodules,
which
small,
rounded
opacities
measuring
8–30
mm.
While
the
concern
about
nodules
is
that
they
may
represent
early
cancer,
programs,
only
1%
such
turn
out
to
be
cancer.
This
leads
series
complex
decisions
and,
at
times,
unnecessary
biopsies
ultimately
determined
benign.
Additionally,
patients
anxious
status
detected
nodules.
The
high
rate
false
positive
nodule
detections
has
driven
advancements
biomarker-based
research
aimed
triaging
(benign
versus
malignant)
identify
truly
malignant
better.
Biomarkers
found
biofluids
and
breath
hold
promise
owing
their
minimally
invasive
sampling
methods,
ease
use,
cost-effectiveness.
Although
several
biomarkers
have
demonstrated
utility,
sensitivity
specificity
still
relatively
low.
Combining
multiple
could
enhance
characterisation
small
pulmonary
by
addressing
limitations
individual
biomarkers.
approach
help
reduce
procedures
accelerate
diagnosis
future.
review
offers
thorough
overview
emerging
emphasising
key
challenges
proposing
potential
solutions
differentiation.
It
focuses
on
rather
than
screening,
analysing
published
primarily
past
five
years
with
some
exceptions.
incorporation
into
will
facilitate
detection
leading
timely
interventions
improved
outcomes.
Further
efforts
needed
increase
cost-effectiveness
practicality
many
applications
settings.
However,
technologies
advancing
rapidly,
soon
implemented
clinics
near
Graphical
abstract
Cancers,
Год журнала:
2023,
Номер
15(17), С. 4344 - 4344
Опубликована: Авг. 30, 2023
Lung
cancer
has
one
of
the
worst
morbidity
and
fatality
rates
any
malignant
tumour.
Most
lung
cancers
are
discovered
in
middle
late
stages
disease,
when
treatment
choices
limited,
patients’
survival
rate
is
low.
The
aim
screening
identification
malignancies
early
stage
more
options
for
effective
treatments
available,
to
improve
outcomes.
desire
efficacy
efficiency
clinical
care
continues
drive
multiple
innovations
into
practice
better
patient
management,
this
context,
artificial
intelligence
(AI)
plays
a
key
role.
AI
may
have
role
each
process
workflow.
First,
acquisition
low-dose
computed
tomography
programs,
AI-based
reconstruction
allows
further
dose
reduction,
while
still
maintaining
an
optimal
image
quality.
can
help
personalization
programs
through
risk
stratification
based
on
collection
analysis
huge
amount
imaging
data.
A
computer-aided
detection
(CAD)
system
provides
automatic
potential
nodules
with
high
sensitivity,
working
as
concurrent
or
second
reader
reducing
time
needed
interpretation.
Once
nodule
been
detected,
it
should
be
characterized
benign
malignant.
Two
approaches
available
perform
task:
first
represented
by
segmentation
consequent
assessment
lesion
size,
volume,
densitometric
features;
consists
first,
followed
radiomic
features
extraction
characterize
whole
abnormalities
providing
so-called
“virtual
biopsy”.
This
narrative
review
aims
provide
overview
all
possible
applications
screening.
Japanese Journal of Radiology,
Год журнала:
2023,
Номер
42(2), С. 190 - 200
Опубликована: Сен. 15, 2023
Abstract
Purpose
In
this
preliminary
study,
we
aimed
to
evaluate
the
potential
of
generative
pre-trained
transformer
(GPT)
series
for
generating
radiology
reports
from
concise
imaging
findings
and
compare
its
performance
with
radiologist-generated
reports.
Methods
This
retrospective
study
involved
28
patients
who
underwent
computed
tomography
(CT)
scans
had
a
diagnosed
disease
typical
findings.
Radiology
were
generated
using
GPT-2,
GPT-3.5,
GPT-4
based
on
patient’s
age,
gender,
site,
We
calculated
top-1,
top-5
accuracy,
mean
average
precision
(MAP)
differential
diagnoses
GPT-4,
radiologists.
Two
board-certified
radiologists
evaluated
grammar
readability,
image
findings,
impression,
diagnosis,
overall
quality
all
4-point
scale.
Results
Top-1
Top-5
accuracies
different
highest
radiologists,
followed
by
in
that
order
(Top-1:
1.00,
0.54,
0.21,
respectively;
Top-5:
0.96,
0.89,
respectively).
There
no
significant
differences
qualitative
scores
about
between
GPT-3.5
or
(
p
>
0.05).
However,
GPT
impression
diagnosis
significantly
lower
than
those
<
Conclusions
Our
suggests
have
possibility
generate
high
readability
reasonable
very
short
keywords;
however,
concerns
persist
regarding
accuracy
impressions
diagnoses,
thereby
requiring
verification
Diagnostic and Interventional Imaging,
Год журнала:
2024,
Номер
105(11), С. 453 - 459
Опубликована: Июнь 24, 2024
The
rapid
advancement
of
artificial
intelligence
(AI)
in
healthcare
has
revolutionized
the
industry,
offering
significant
improvements
diagnostic
accuracy,
efficiency,
and
patient
outcomes.
However,
increasing
adoption
AI
systems
also
raises
concerns
about
their
environmental
impact,
particularly
context
climate
change.
This
review
explores
intersection
change
healthcare,
examining
challenges
posed
by
energy
consumption
carbon
footprint
systems,
as
well
potential
solutions
to
mitigate
impact.
highlights
energy-intensive
nature
model
training
deployment,
contribution
data
centers
greenhouse
gas
emissions,
generation
electronic
waste.
To
address
these
challenges,
development
energy-efficient
models,
green
computing
practices,
integration
renewable
sources
are
discussed
solutions.
emphasizes
role
optimizing
workflows,
reducing
resource
waste,
facilitating
sustainable
practices
such
telemedicine.
Furthermore,
importance
policy
governance
frameworks,
global
initiatives,
collaborative
efforts
promoting
is
explored.
concludes
outlining
best
for
including
eco-design,
lifecycle
assessment,
responsible
management,
continuous
monitoring
improvement.
As
industry
continues
embrace
technologies,
prioritizing
sustainability
responsibility
crucial
ensure
that
benefits
realized
while
actively
contributing
preservation
our
planet.
In
the
context
of
rapid
technological
advancements,
narrative
review
titled
"Digital
Pathology:
Transforming
Diagnosis
in
Digital
Age"
explores
significant
impact
digital
pathology
reshaping
diagnostic
approaches.
This
delves
into
various
effects
field,
including
remote
consultations
and
artificial
intelligence
(AI)-assisted
analysis,
revealing
ongoing
transformation
taking
place.
The
investigation
process
digitizing
traditional
glass
slides,
which
aims
to
improve
accessibility
facilitate
sharing.
Additionally,
it
addresses
complexities
associated
with
data
security
standardization
challenges.
Incorporating
AI
enhances
pathologists'
capabilities
accelerates
analytical
procedures.
Furthermore,
highlights
growing
importance
collaborative
networks
facilitating
global
knowledge
It
also
emphasizes
this
technology
on
medical
education
patient
care.
provide
an
overview
pathology's
transformative
innovative
potential,
highlighting
its
disruptive
nature
practices.
BMC Medical Imaging,
Год журнала:
2023,
Номер
23(1)
Опубликована: Сен. 15, 2023
Abstract
Background
Vision
transformer-based
methods
are
advancing
the
field
of
medical
artificial
intelligence
and
cancer
imaging,
including
lung
applications.
Recently,
many
researchers
have
developed
vision
AI
for
diagnosis
prognosis.
Objective
This
scoping
review
aims
to
identify
recent
developments
on
imaging
It
provides
key
insights
into
how
transformers
complemented
performance
deep
learning
cancer.
Furthermore,
also
identifies
datasets
that
contributed
field.
Methods
In
this
review,
we
searched
Pubmed,
Scopus,
IEEEXplore,
Google
Scholar
online
databases.
The
search
terms
included
intervention
(vision
transformers)
task
(i.e.,
cancer,
adenocarcinoma,
etc.).
Two
reviewers
independently
screened
title
abstract
select
relevant
studies
performed
data
extraction.
A
third
reviewer
was
consulted
validate
inclusion
exclusion.
Finally,
narrative
approach
used
synthesize
data.
Results
Of
314
retrieved
studies,
34
published
from
2020
2022.
most
commonly
addressed
in
these
classification
types,
such
as
squamous
cell
carcinoma
versus
identifying
benign
malignant
pulmonary
nodules.
Other
applications
survival
prediction
patients
segmentation
lungs.
lacked
clear
strategies
clinical
transformation.
SWIN
transformer
a
popular
choice
researchers;
however,
other
architectures
were
reported
where
combined
with
convolutional
neural
networks
or
UNet
model.
Researchers
publicly
available
database
consortium
genome
atlas.
One
study
cluster
48
GPUs,
while
one,
two,
four
GPUs.
Conclusion
can
be
concluded
models
increasingly
popularity
developing
However,
their
computational
complexity
relevance
important
factors
considered
future
research
work.
valuable
healthcare
advance
state-of-the-art
We
provide
an
interactive
dashboard
lung-cancer.onrender.com/
.
Magnetic Resonance in Medical Sciences,
Год журнала:
2023,
Номер
22(4), С. 401 - 414
Опубликована: Янв. 1, 2023
Due
primarily
to
the
excellent
soft
tissue
contrast
depictions
provided
by
MRI,
widespread
application
of
head
and
neck
MRI
in
clinical
practice
serves
assess
various
diseases.
Artificial
intelligence
(AI)-based
methodologies,
particularly
deep
learning
analyses
using
convolutional
neural
networks,
have
recently
gained
global
recognition
been
extensively
investigated
research
for
their
applicability
across
a
range
categories
within
medical
imaging,
including
MRI.
Analytical
approaches
AI
shown
potential
addressing
limitations
associated
with
In
this
review,
we
focus
on
technical
advancements
deep-learning-based
methodologies
utility
field
encompassing
aspects
such
as
image
acquisition
reconstruction,
lesion
segmentation,
disease
classification
diagnosis,
prognostic
prediction
patients
presenting
We
then
discuss
current
offer
insights
regarding
future
challenges
field.
