The British Journal of Psychiatry,
Год журнала:
2024,
Номер
unknown, С. 1 - 6
Опубликована: Ноя. 5, 2024
Background
Attempts
to
use
artificial
intelligence
(AI)
in
psychiatric
disorders
show
moderate
success,
highlighting
the
potential
of
incorporating
information
from
clinical
assessments
improve
models.
This
study
focuses
on
using
large
language
models
(LLMs)
detect
suicide
risk
medical
text
care.
Aims
To
extract
about
suicidality
status
admission
notes
electronic
health
records
(EHRs)
privacy-sensitive,
locally
hosted
LLMs,
specifically
evaluating
efficacy
Llama-2
Method
We
compared
performance
several
variants
open
source
LLM
extracting
100
reports
against
a
ground
truth
defined
by
human
experts,
assessing
accuracy,
sensitivity,
specificity
and
F1
score
across
different
prompting
strategies.
Results
A
German
fine-tuned
model
showed
highest
accuracy
(87.5%),
sensitivity
(83.0%)
(91.8%)
identifying
suicidality,
with
significant
improvements
various
prompt
designs.
Conclusions
The
demonstrates
capability
particularly
Llama-2,
accurately
while
preserving
data
privacy.
suggests
their
application
surveillance
systems
for
emergencies
improving
management
systematic
quality
control
research.
JAMA Psychiatry,
Год журнала:
2024,
Номер
81(4), С. 386 - 386
Опубликована: Янв. 10, 2024
Biological
psychiatry
aims
to
understand
mental
disorders
in
terms
of
altered
neurobiological
pathways.
However,
for
one
the
most
prevalent
and
disabling
disorders,
major
depressive
disorder
(MDD),
no
informative
biomarkers
have
been
identified.
Human Brain Mapping,
Год журнала:
2025,
Номер
46(1)
Опубликована: Янв. 1, 2025
ABSTRACT
Trait
mindfulness
refers
to
one's
disposition
or
tendency
pay
attention
their
experiences
in
the
present
moment,
a
non‐judgmental
and
accepting
way.
has
been
robustly
associated
with
positive
mental
health
outcomes,
but
its
neural
underpinnings
are
poorly
understood.
Prior
resting‐state
fMRI
studies
have
trait
within‐
between‐network
connectivity
of
default‐mode
(DMN),
fronto‐parietal
(FPN),
salience
networks.
However,
it
is
unclear
how
generalizable
findings
are,
they
relate
different
components
mindfulness,
other
networks
brain
areas
may
be
involved.
To
address
these
gaps,
we
conducted
largest
study
to‐date,
consisting
pre‐registered
connectome‐based
predictive
modeling
analysis
367
meditation‐naïve
adults
across
three
samples
collected
at
sites.
In
model‐training
dataset,
did
not
find
connections
that
predicted
overall
identified
models
two
subscales,
Acting
Awareness
Non‐judging
.
Models
included
both
(sets
pairwise
positively
increasing
connectivity)
negative
networks,
which
showed
inverse
relationship.
The
network
distinct
representations
involving
FPN
DMN,
respectively.
models,
overlapped
significantly
involved
whole
prominent
involvement
somatomotor,
visual
DMN
Only
generalized
predict
subscale
scores
out‐of‐sample,
test
datasets.
Predictions
from
were
also
negatively
correlated
predictions
well‐established
mind‐wandering
connectome
model.
We
preliminary
evidence
for
based
on
specific
affective
cognitive
facets.
incomplete
generalization
all
sites
scanners,
limited
stability
as
well
substantial
overlap
between
underscores
difficulty
finding
robust
markers
Biomedicines,
Год журнала:
2025,
Номер
13(1), С. 167 - 167
Опубликована: Янв. 12, 2025
Background/Objectives:
The
dual
forces
of
structured
inquiry
and
serendipitous
discovery
have
long
shaped
neuropsychiatric
research,
with
groundbreaking
treatments
such
as
lithium
ketamine
resulting
from
unexpected
discoveries.
However,
relying
on
chance
is
becoming
increasingly
insufficient
to
address
the
rising
prevalence
mental
health
disorders
like
depression
schizophrenia,
which
necessitate
precise,
innovative
approaches.
Emerging
technologies
artificial
intelligence,
induced
pluripotent
stem
cells,
multi-omics
potential
transform
this
field
by
allowing
for
predictive,
patient-specific
interventions.
Despite
these
advancements,
traditional
methodologies
animal
models
single-variable
analyses
continue
be
used,
frequently
failing
capture
complexities
human
conditions.
Summary:
This
review
critically
evaluates
transition
serendipity
precision-based
in
research.
It
focuses
key
innovations
dynamic
systems
modeling
network-based
approaches
that
use
genetic,
molecular,
environmental
data
identify
new
therapeutic
targets.
Furthermore,
it
emphasizes
importance
interdisciplinary
collaboration
human-specific
overcoming
limitations
Conclusions:
We
highlight
precision
psychiatry’s
transformative
revolutionizing
care.
paradigm
shift,
combines
cutting-edge
systematic
frameworks,
promises
increased
diagnostic
accuracy,
reproducibility,
efficiency,
paving
way
tailored
better
patient
outcomes
JAMA Psychiatry,
Год журнала:
2025,
Номер
82(3), С. 215 - 215
Опубликована: Янв. 22, 2025
Our
website
uses
cookies
to
enhance
your
experience.
By
continuing
use
our
site,
or
clicking
"Continue,"
you
are
agreeing
Cookie
Policy
|
Continue
JAMA
Psychiatry
HomeNew
OnlineCurrent
IssueFor
Authors
Podcast
JAMA+
AI
Journals
Network
Open
Cardiology
Dermatology
Health
Forum
Internal
Medicine
Neurology
Oncology
Ophthalmology
Otolaryngology–Head
&
Neck
Surgery
Pediatrics
Archives
of
(1919-1959)
JN
Learning
/
CMESubscribeJobsInstitutions
LibrariansReprints
Permissions
Terms
Use
Privacy
Accessibility
Statement
2025
American
Medical
Association.
All
Rights
Reserved
Search
Archive
Input
Term
Sign
In
Individual
inCreate
an
Account
Access
through
institution
Purchase
Options:
Buy
this
article
Rent
Subscribe
the
journal
Applied Sciences,
Год журнала:
2025,
Номер
15(3), С. 1132 - 1132
Опубликована: Янв. 23, 2025
Reliably
detecting
COVID-19
is
critical
for
diagnosis
and
disease
control.
However,
imbalanced
data
in
medical
datasets
pose
significant
challenges
machine
learning
models,
leading
to
bias
poor
generalization.
The
dataset
obtained
from
the
EPIVIGILA
system
Chilean
Epidemiological
Surveillance
Process
contains
information
on
over
6,000,000
patients,
but,
like
many
current
datasets,
it
suffers
class
imbalance.
To
address
this
issue,
we
applied
various
algorithms,
both
with
without
sampling
methods,
compared
them
using
different
classification
diagnostic
metrics
such
as
precision,
sensitivity,
specificity,
likelihood
ratio
positive,
odds
ratio.
Our
results
showed
that
applying
methods
improved
metric
values
contributed
models
better
Effectively
managing
crucial
reliable
diagnosis.
This
study
enhances
understanding
of
how
techniques
can
improve
reliability
contribute
patient
outcomes.
