Frontiers in Immunology,
Год журнала:
2024,
Номер
15
Опубликована: Март 12, 2024
Mechanistic
learning
refers
to
the
synergistic
combination
of
mechanistic
mathematical
modeling
and
data-driven
machine
or
deep
learning.
This
emerging
field
finds
increasing
applications
in
(mathematical)
oncology.
review
aims
capture
current
state
provides
a
perspective
on
how
may
progress
oncology
domain.
We
highlight
potential
point
out
similarities
differences
between
purely
approaches
concerning
model
complexity,
data
requirements,
outputs
generated,
interpretability
algorithms
their
results.
Four
categories
(sequential,
parallel,
extrinsic,
intrinsic)
are
presented
with
specific
examples.
discuss
range
techniques
including
physics-informed
neural
networks,
surrogate
learning,
digital
twins.
Example
address
complex
problems
predominantly
from
domain
research
such
as
longitudinal
tumor
response
predictions
time-to-event
modeling.
As
advances,
we
aim
for
this
proposed
categorization
framework
foster
additional
collaboration
data-
knowledge-driven
fields.
Further
will
help
difficult
issues
limited
availability,
requirements
transparency,
input
which
embraced
npj Digital Medicine,
Год журнала:
2024,
Номер
7(1)
Опубликована: Июль 16, 2024
Virtual
patients
and
digital
patients/twins
are
two
similar
concepts
gaining
increasing
attention
in
health
care
with
goals
to
accelerate
drug
development
improve
patients'
survival,
but
their
own
limitations.
Although
methods
have
been
proposed
generate
virtual
patient
populations
using
mechanistic
models,
there
limited
number
of
applications
immuno-oncology
research.
Furthermore,
due
the
stricter
requirements
twins,
they
often
generated
a
study-specific
manner
models
customized
particular
clinical
settings
(e.g.,
treatment,
cancer,
data
types).
Here,
we
discuss
challenges
for
generation
our
most
recent
experiences,
initiatives
develop
how
research
on
these
can
inform
each
other.
Journal of Translational Medicine,
Год журнала:
2025,
Номер
23(1)
Опубликована: Март 19, 2025
As
global
cancer
incidence
and
mortality
rise,
digital
twin
technology
in
precision
medicine
offers
new
opportunities
for
treatment.
This
study
aims
to
systematically
analyze
the
current
applications,
research
trends,
challenges
of
tumor
therapy,
while
exploring
future
directions.
Relevant
literature
up
2024
was
retrieved
from
PubMed,
Web
Science,
other
databases.
Data
visualization
performed
using
R
VOSviewer
software.
The
analysis
includes
initiation
funding
models,
distribution,
sample
size
analysis,
data
processing
artificial
intelligence
applications.
Furthermore,
investigates
specific
applications
effectiveness
diagnosis,
treatment
decision-making,
prognosis
prediction,
personalized
management.
Since
2020,
on
oncology
has
surged,
with
significant
contributions
United
States,
Germany,
Switzerland,
China.
Funding
primarily
comes
government
agencies,
particularly
National
Institutes
Health
U.S.
Sample
reveals
that
large-sample
studies
have
greater
clinical
reliability,
small-sample
emphasize
validation.
In
integration
medical
imaging,
multi-omics
data,
AI
algorithms
is
key.
By
combining
multimodal
dynamic
modeling,
accuracy
models
been
significantly
improved.
However,
different
types
still
faces
related
tool
interoperability
limited
standardization.
Specific
shown
advantages
surgical
planning.
Digital
holds
substantial
promise
therapy
by
optimizing
plans
through
integrated
modeling.
factors
such
as
language
restrictions,
potential
selection
bias,
relatively
small
number
published
this
emerging
field,
which
may
affect
comprehensiveness
generalizability
our
findings.
Moreover,
issues
heterogeneity,
technical
integration,
privacy
ethics
continue
impede
its
broader
application.
Future
should
promote
international
collaboration,
establish
unified
interdisciplinary
standards,
strengthen
ethical
regulations
accelerate
translation
Mathematics,
Год журнала:
2024,
Номер
12(8), С. 1195 - 1195
Опубликована: Апрель 16, 2024
This
paper
presents
a
comprehensive
investigation
into
the
applicability
and
performance
of
two
prominent
growth
models,
namely,
Verhulst
model
Montroll
model,
in
context
modeling
tumor
cell
dynamics.
Leveraging
power
Physics-Informed
Neural
Networks
(PINNs),
we
aim
to
assess
compare
predictive
capabilities
these
models
against
experimental
data
obtained
from
patterns
cells.
We
employed
dataset
comprising
detailed
measurements
train
evaluate
models.
By
integrating
PINNs,
not
only
account
for
noise
but
also
embed
physical
insights
learning
process,
enabling
capture
underlying
mechanisms
governing
growth.
Our
findings
reveal
strengths
limitations
each
accurately
representing
proliferation
Furthermore,
study
sheds
light
on
impact
incorporating
physics-informed
constraints
predictions.
The
gained
this
comparative
analysis
contribute
advancing
our
understanding
their
applications
predicting
complex
biological
phenomena,
particularly
realm
proliferation.
Journal of Chemometrics,
Год журнала:
2025,
Номер
39(2)
Опубликована: Фев. 1, 2025
ABSTRACT
The
analysis
of
single
cell
multi‐omics
data
is
a
complex
task,
and
many
explorative
methods
are
being
used
to
draw
information
from
such
data.
This
paper
compares
several
these
visualize
the
output
mechanistic
model
under
various
simulated
conditions.
include
PCA,
PARAFAC,
ASCA,
MASCARA,
COVSCA,
P‐ESCA,
PE‐ASCA.
These
techniques,
applied
high‐dimensional
as
gene
expression
protein
levels,
assess
correlations
across
time
series
experimental
study
uses
MCF10A
cancer
cells,
simulating
interactions
between
signaling
pathways
related
growth
division.
Results
show
that
while
like
PCA
PARAFAC
ASCA
reveal
time‐dependent
variations
in
data,
mRNA
exhibit
minimal
systematic
variation.
MASCARA
offers
unique
insights
by
identifying
genes
linked
specific
pathways.
work
highlights
potential
limitations
understanding
particularly
single‐cell
contexts
where
variation
stochastic
processes
complicate
interpretation.
Multidisciplinary Reviews,
Год журнала:
2025,
Номер
8(7), С. 2025218 - 2025218
Опубликована: Фев. 7, 2025
In
the
area
of
theragnostics,
use
artificial
intelligence
(AI)
is
supporting
personalised
medicine
methods
that
merge
therapeutic
and
diagnostic
techniques,
which
causing
sector
to
undergo
a
transition.
An
analysis
historical
backdrop,
current
condition,
promise
intelligence-enhanced
theragnostic
systems
presented
in
this
article.
We
investigate
underlying
ideas
intelligence,
such
as
machine
learning,
deep
neural
networks,
well
their
applications
variety
medical
fields,
including
cancer,
pathology,
imaging,
cardiology,
hypertension
control,
diabetes
management.
The
ability
integrate
wide
information,
recognise
trends,
enable
real-time
decision-making
patient
monitoring
all
illustrate
competency.
It
possible
digital
twins,
make
adaptive
learning
algorithms
dynamic
virtual
models,
might
be
used
optimise
treatment
regimens
anticipate
course
illness.
Important
prospects
for
advancement
biomedical
research
therapy
are
by
biochip
technology
driven
intelligence.
This
includes
gene
chips,
organ-on-a-chip
systems,
biosensors.
However,
there
number
obstacles
must
overcome
before
can
effectively
theragnostics.
These
include
data
security,
privacy,
algorithmic
biases,
legal
frameworks,
acceptability.
vital,
order
realise
full
potential
AI-driven
address
these
constraints
means
extensive
validation,
diversified
datasets,
explainable
clear
communication.
anticipated
synergistic
combination
theragnostics
will
revolutionise
precision
continues
advance.
it
more
accurate
diagnoses,
achieve
tailored
therapeutics,
better
outcomes.
Given
the
enormous
output
and
pace
of
development
artificial
intelligence
(AI)
methods
in
medical
imaging,
it
can
be
challenging
to
identify
true
success
stories
determine
state-of-the-art
field.
This
report
seeks
provide
magnetic
resonance
imaging
(MRI)
community
with
an
initial
guide
into
major
areas
which
AI
are
contributing
MRI
oncology.
After
a
general
introduction
intelligence,
we
proceed
discuss
successes
current
limitations
when
used
for
image
acquisition,
reconstruction,
registration,
segmentation,
as
well
its
utility
assisting
diagnostic
prognostic
settings.
Within
each
section,
attempt
present
balanced
summary
by
first
presenting
common
techniques,
state
readiness,
clinical
needs,
barriers
practical
deployment
setting.
We
conclude
new
advances
must
realized
address
questions
regarding
generalizability,
quality
assurance
control,
uncertainty
quantification
applying
cancer
maintain
patient
safety
utility.
Frontiers in Materials,
Год журнала:
2025,
Номер
12
Опубликована: Апрель 9, 2025
Cancer
is
one
of
the
deadliest
diseases
despite
aggressive
therapeutics.
This
due
in
part
to
evolving
tumor
microenvironment
(TME),
which
provide
supportive
cues
that
promote
adaptation
and
progression.
Emerging
studies
highlight
significant
role
biophysical
characteristics
TME
modulating
all
aspects
cancer
spread.
With
advance
bioengineering
platforms,
deeper
investigations
into
impact
these
features
on
progression
are
being
conducted
with
a
growing
appreciation
intratumoral
compression
underlie
many
changes.
Intratumoral
emerges
early
development
increases
magnitude
as
rapidly
expands
against
itself
its
surrounding
tissue.
stress
has
effects
both
cells
TME,
including
hypoxia,
shear
stress,
extracellular
matrix
(ECM)
remodeling,
substrate
stiffness.
creates
physically
dense,
pro-malignant
environment
can
metastatic
phenotypes
spread
but
also
present
barriers
for
immune
cell
infiltration.
review
will
analyze
effect
compressive
cells,
confined
migration
populations.
