Background:
Ovarian,
cervical,
and
endometrial
cancers
stand
as
fatal
health
killers
for
women's
mental
physical
health,
especially
affecting
female
reproductive
systems.
Exploiting
weakly
supervised
learning
explainable
AI
techniques
are
crucial
fast,
accurate,
robust
automatic
marker
detection,
efficient
prevention,
primary
treatment
of
gynecological
tumors.
Methods:
With
respect
to
the
PRIMSA
methodology,
schemes
deep
learning-based
investigated
in
cross-subject
fields
clinical
diagnostic
imaging
technology.
Related
methods,
opening
research
problems,
challenging
subjects
explored
screening
tumors
performing
cancer
image
diagnosis
latest
study.
Results:
Keynote
approaches
combining
ultrasound
medicine,
AI,
technology
summarized.
In
combination
with
methods
medical
imaging,
prospective
insights
put
forward
refreshed
concepts
on
treatments
area
oncology,
a
feasible
range
applications
their
corresponding
techniques.
Conclusions:
Explainable
capable
accurate
classification
between
benign
malignant
tumors,
yielding
pathway
care
improving
survival
rate
patients,
pacing
disease
prediction,
matching
strategic
goal
"Healthy
China
2030".
Journal of Clinical Medicine,
Journal Year:
2024,
Volume and Issue:
13(2), P. 610 - 610
Published: Jan. 21, 2024
Thymomas
are
rare
tumors
of
the
anterior
mediastinum
with
peculiar
clinical
and
pathological
features.
They
have
been
deeply
analyzed
by
pioneer
authors,
who
strictly
linked
their
name
to
main
staging
classifications.
Before
latest
edition
WHO
classification
thymic
epithelial
tumors,
history
thymoma
inherited
pathologists
systematically
addressed
issue,
from
Levine-Rosai
Muller-Hermelink.
Similarly,
system
is
intimately
related
two
surgeons,
Masaoka
Koga,
historically
dealt
this
disease.
More
recently,
traditional
tumor-nodes-metastasis
(TNM)
has
developed
for
condition,
in
a
rational
attempt
put
thymomas
conformity
other
solid
tumors.
The
efforts
International
Thymic
Malignancies
Interest
Group
(ITMIG)
Domain
Staging
Prognostic
Factors
Committee
(TD-SPFC)
Association
Study
Lung
Cancer
(IASLC)
resulted
TNM
which
included
eighth
American
Joint
on
Cancer’s
(AJCC)
Manual.
Herein,
we
report
narrative
review
evolution
(TET)
present
critical
appraisal
actual
compared
historical
Masaoka-Koga
classification,
special
focus
proposal
ninth
TNM,
expected
2024.
Heliyon,
Journal Year:
2024,
Volume and Issue:
10(7), P. e29249 - e29249
Published: April 1, 2024
Peritoneal
carcinomatosis
(PC)
is
a
type
of
secondary
cancer
which
not
sensitive
to
conventional
intravenous
chemotherapy.
Treatment
strategies
for
PC
are
usually
palliative
rather
than
curative.
Recently,
artificial
intelligence
(AI)
has
been
widely
used
in
the
medical
field,
making
early
diagnosis,
individualized
treatment,
and
accurate
prognostic
evaluation
various
cancers,
including
mediastinal
malignancies,
colorectal
cancer,
lung
more
feasible.
As
branch
computer
science,
AI
specializes
image
recognition,
speech
automatic
large-scale
data
extraction
output.
technologies
have
also
made
breakthrough
progress
field
peritoneal
based
on
its
powerful
learning
capacity
efficient
computational
power.
successfully
applied
approaches
imaging,
blood
tests,
proteomics,
pathological
diagnosis.
Due
function
convolutional
neural
network
model
machine
algorithms,
AI-assisted
diagnosis
types
associated
with
higher
accuracy
rate
compared
methods.
In
addition,
treatment
surgical
resection,
intraperitoneal
chemotherapy,
systemic
significantly
improves
survival
patients
PC.
particular,
recurrence
prediction
emotion
combined
technology,
further
improving
quality
life
patients.
Here
we
comprehensively
reviewed
summarized
latest
developments
application
PC,
helping
oncologists
diagnose
provide
precise
New Directions for Evaluation,
Journal Year:
2023,
Volume and Issue:
2023(178-179), P. 23 - 32
Published: June 1, 2023
Abstract
In
this
article,
I
discuss
the
use
of
artificial
intelligence
(AI)
in
evaluation
and
its
relevance
to
evolution
field.
begin
with
a
background
on
how
AI
models
are
developed,
including
machine
learning
makes
sense
data
algorithms
it
develops
go
power
models.
explain
foundational
understanding
natural
language
processing
informs
where
might
not
be
effectively
used.
A
critical
concern
is
that
only
as
strong
which
they
trained,
evaluators
should
consider
important
limitations
when
using
AI,
structural
inequality.
considering
relationship
between
evaluation,
must
both
AI's
an
evaluative
tool
role
new
subject
evaluation.
As
becomes
more
relevant
wider
array
fields
disciplines,
will
need
develop
strategies
for
good
(or
not),
what
not)
do.
Deleted Journal,
Journal Year:
2024,
Volume and Issue:
37(3), P. 1187 - 1200
Published: Feb. 8, 2024
Abstract
Segmentation
and
image
intensity
discretization
impact
on
radiomics
workflow.
The
aim
of
this
study
is
to
investigate
the
influence
interobserver
segmentation
variability
methods
reproducibility
MRI-based
radiomic
features
in
lipoma
atypical
lipomatous
tumor
(ALT).
Thirty
patients
with
or
ALT
were
retrospectively
included.
Three
readers
independently
performed
manual
contour-focused
T1-weighted
T2-weighted
sequences,
including
whole
volume.
Additionally,
a
marginal
erosion
was
applied
segmentations
evaluate
its
feature
reproducibility.
After
pre-processing,
included
employing
both
fixed
bin
number
width
approaches,
1106
extracted
from
each
sequence.
Intraclass
correlation
coefficient
(ICC)
95%
confidence
interval
lower
bound
≥
0.75
defined
stability.
In
vs.
margin
shrinkage
segmentation,
rates
stable
images
ranged
92.68
95.21%
90.69
95.66%
after
95.75
97.65%
95.39
96.47%
discretization,
respectively,
no
difference
between
two
approaches
(
p
0.175).
Higher
higher
ICC
values
found
when
implementing
compared
number,
regardless
approach
<
0.001).
conclusion,
MRI
are
reproducible
method,
although
certain
degree
highlights
need
for
preliminary
reliability
analysis
future
studies.
Journal of Translational Medicine,
Journal Year:
2024,
Volume and Issue:
22(1)
Published: June 13, 2024
Both
cancer
and
fibrosis
are
diseases
involving
dysregulation
of
cell
signaling
pathways
resulting
in
an
altered
cellular
microenvironment
which
ultimately
leads
to
progression
the
condition.
The
two
disease
entities
share
common
molecular
pathophysiology
recent
research
has
illuminated
how
each
promotes
other.
Multiple
imaging
techniques
have
been
developed
aid
early
accurate
diagnosis
disease,
given
commonalities
between
conditions,
advances
one
opened
new
avenues
study
Here,
we
detail
most
up-to-date
for
they
crossed
over
improve
detection
monitoring
We
explore
positron
emission
tomography
(PET),
magnetic
resonance
(MRI),
second
generation
harmonic
Imaging
(SGHI),
ultrasound
(US),
radiomics,
artificial
intelligence
(AI).
A
diagnostic
tool
PET/computed
(CT)
is
use
radiolabeled
fibroblast
activation
protein
inhibitor
(FAPI).
SGHI
uses
high-frequency
sound
waves
penetrate
deeper
into
tissue,
providing
a
more
detailed
view
tumor
microenvironment.
Artificial
with
advanced
deep
learning
(DL)
algorithms
highly
effective
training
computer
systems
diagnose
classify
neoplastic
lesions
multiple
organs.
Ultimately,
advancing
can
lead
significantly
timely
diagnoses
both
better
patient
outcomes.
Cancers,
Journal Year:
2024,
Volume and Issue:
16(10), P. 1868 - 1868
Published: May 14, 2024
We
aimed
to
analyze
the
management
of
ectopic
mediastinal
thyroid
(EMT)
with
respect
EMT-related
cancer
and
non-malignant
findings
related
pathological
report,
clinical
presentation,
imaging
traits,
endocrine
profile,
connective
tissue
cervical
(eutopic)
gland,
biopsy
or
fine
needle
aspiration
(FNA)
results,
surgical
techniques
post-operatory
outcome.
This
was
a
comprehensive
review
based
on
revising
any
type
freely
PubMed-accessible
English,
full-length
original
papers
including
keywords
“ectopic
thyroid”
“mediastinum”
from
inception
until
March
2024.
included
89
articles
that
specified
EMTs
data.
classified
them
into
four
main
groups:
(I)
studies/case
series
(n
=
10;
N
36
EMT
patients);
(II)
malignant
(N
22
subjects;
except
for
one
newborn
immature
teratoma
in
EMT,
only
adults
were
reported;
mean
age
62.94
years;
ranges:
34
90
female
male
ratio
0.9).
Histological
analysis
showed
following:
papillary
11/21);
follicular
variant
2/21);
Hürthle
cell
malignancy
1/21);
poorly
differentiated
anaplastic
medullary
lymphoma
MALT
(mucosa-associated
lymphoid
tissue)
(III)
benign
no
anomalies
37
56.32
30
80
1.8);
(IV)
23;
5.6;
average
52.1
years).
panel
involved
clinical/subclinical
hypothyroidism
(iatrogenic,
congenital,
thyroiditis-induced,
transitory
upon
removal);
thyrotoxicosis
(including
autonomous
activity
suppressed
eutopic
gland);
autoimmune
thyroiditis/Graves’s
disease;
nodules/multinodular
goiter
prior
thyroidectomy
(before
detection).
propose
10-item
algorithm
might
help
navigate
through
domain.
To
conclude,
across
this
focused-sample
(to
our
knowledge,
largest
its
kind)
EMTs,
index
suspicion
remains
low;
higher
rate
is
reported
than
data
(18.8%),
incident
imagery-based
detection
found
10–14%
EMTs;
surgery
offered
an
overall
good
A
wide
range
imagery,
biopsy/FNA
procedures
part
otherwise
complex
personalized
management.
