Materials,
Год журнала:
2021,
Номер
14(9), С. 2374 - 2374
Опубликована: Май 2, 2021
In
this
review,
we
present
an
overview
of
significant
developments
in
the
field
situ
and
operando
(ISO)
X-ray
imaging
solidification
processes.
The
objective
review
is
to
emphasize
key
challenges
developing
performing
processes,
as
well
highlight
important
contributions
that
have
significantly
advanced
understanding
various
mechanisms
pertaining
microstructural
evolution,
defects,
semi-solid
deformation
metallic
alloy
systems.
Likewise,
some
process
modifications
such
electromagnetic
ultra-sound
melt
treatments
also
been
described.
Finally,
a
discussion
on
recent
breakthroughs
emerging
technology
additive
manufacturing,
thereof,
are
presented.
Journal of Medical Radiation Sciences,
Год журнала:
2023,
Номер
71(2), С. 290 - 298
Опубликована: Окт. 4, 2023
Automation
and
artificial
intelligence
(AI)
is
already
possible
for
many
radiation
therapy
planning
treatment
processes
with
the
aim
of
improving
workflows
increasing
efficiency
in
oncology
departments.
Currently,
AI
technology
advancing
at
an
exponential
rate,
as
are
its
applications
oncology.
This
commentary
highlights
way
has
begun
to
impact
looks
ahead
potential
future
developments
this
space.
Historically,
therapist's
(RT's)
role
evolved
alongside
adoption
new
technology.
In
Australia,
RTs
have
key
clinical
roles
both
delivery
been
integral
implementation
automated
solutions
areas.
They
will
need
continue
be
informed,
adapt
transform
technologies
implemented
into
practice
play
important
how
AI-based
automation
ensuring
application
can
truly
enable
personalised
higher-quality
patients.
To
inform
optimise
utilisation
AI,
research
should
not
only
focus
on
outcomes
but
also
AI's
professional
roles,
responsibilities
service
delivery.
Increased
efficiencies
workflow
workforce
maintain
safe
improvements
come
cost
creativity,
innovation,
oversight
safety.
Insights into Imaging,
Год журнала:
2023,
Номер
14(1)
Опубликована: Авг. 25, 2023
This
study
focuses
on
assessing
the
performance
of
active
learning
techniques
to
train
a
brain
MRI
glioma
segmentation
model.The
publicly
available
training
dataset
provided
for
2021
RSNA-ASNR-MICCAI
Brain
Tumor
Segmentation
(BraTS)
Challenge
was
used
in
this
study,
consisting
1251
multi-institutional,
multi-parametric
MR
images.
Post-contrast
T1,
T2,
and
T2
FLAIR
images
as
well
ground
truth
manual
were
input
model.
The
data
split
into
set
1151
cases
testing
100
cases,
with
remaining
constant
throughout.
Deep
convolutional
neural
network
models
trained
using
NiftyNet
platform.
To
test
viability
model,
an
initial
reference
model
all
followed
by
two
additional
only
575
cases.
resulting
predicted
segmentations
these
then
addended
training.It
demonstrated
that
approach
can
lead
comparable
gliomas
(0.906
Dice
score
vs
0.868
score)
while
requiring
annotation
28.6%
data.The
when
applied
drastically
reduce
time
labor
spent
preparation
data.Active
concepts
deep
learning-assisted
from
assess
their
reducing
required
amount
manually
annotated
training.•
•
gliomas.
Active
data.
Materials,
Год журнала:
2021,
Номер
14(9), С. 2374 - 2374
Опубликована: Май 2, 2021
In
this
review,
we
present
an
overview
of
significant
developments
in
the
field
situ
and
operando
(ISO)
X-ray
imaging
solidification
processes.
The
objective
review
is
to
emphasize
key
challenges
developing
performing
processes,
as
well
highlight
important
contributions
that
have
significantly
advanced
understanding
various
mechanisms
pertaining
microstructural
evolution,
defects,
semi-solid
deformation
metallic
alloy
systems.
Likewise,
some
process
modifications
such
electromagnetic
ultra-sound
melt
treatments
also
been
described.
Finally,
a
discussion
on
recent
breakthroughs
emerging
technology
additive
manufacturing,
thereof,
are
presented.
