Evaluating predictive artificial intelligence approaches used in mobile health platforms to forecast mental health symptoms among youth: a systematic review
Psychiatry Research,
Год журнала:
2024,
Номер
343, С. 116277 - 116277
Опубликована: Ноя. 19, 2024
The
youth
mental
health
crisis
is
exacerbated
by
limited
access
to
care
and
resources.
Mobile
(mHealth)
platforms
using
predictive
artificial
intelligence
(AI)
can
improve
reduce
barriers,
enabling
real-time
responses
precision
prevention.
This
systematic
review
evaluates
AI
approaches
in
mHealth
for
forecasting
symptoms
among
(13-25
years).
We
searched
studies
from
Embase,
PubMed,
Web
of
Science,
PsycInfo,
CENTRAL,
identify
relevant
studies.
From
11
identified,
three
predicted
multiple
symptoms,
with
depression
being
the
most
common
(63%).
Most
used
smartphones
25%
integrated
wearables.
Key
predictors
included
smartphone
usage
(N=5),
sleep
metrics
(N=6),
physical
activity
(N=5).
Nuanced
like
locations
stages
improved
prediction.
Logistic
regression
was
followed
Support
Vector
Machines
(N=3)
ensemble
methods
(N=4).
F-scores
anxiety
ranged
0.73
0.84,
AUCs
0.50
0.74.
Stress
models
had
0.68
0.83.
Bayesian
model
selection
Shapley
values
enhanced
robustness
interpretability.
Barriers
small
sample
sizes,
privacy
concerns,
missing
data,
underrepresentation
bias.
Rigorous
evaluation
performance,
generalizability,
user
engagement
critical
before
are
into
psychiatric
care.
Язык: Английский
It's late, but not too late to transform health systems: a global digital citizen science observatory for local solutions to global problems
Frontiers in Digital Health,
Год журнала:
2024,
Номер
6
Опубликована: Ноя. 27, 2024
A
key
challenge
in
monitoring,
managing,
and
mitigating
global
health
crises
is
the
need
to
coordinate
clinical
decision-making
with
systems
outside
of
healthcare.
In
21st
century,
human
engagement
Internet-connected
ubiquitous
devices
generates
an
enormous
amount
big
data,
which
can
be
used
address
complex,
intersectoral
problems
via
participatory
epidemiology
mHealth
approaches
that
operationalized
digital
citizen
science.
These
data
-
traditionally
exist
are
underutilized
even
though
their
usage
have
significant
implications
for
prediction
prevention
communicable
non-communicable
diseases.
To
critical
challenges
gaps
utilization
across
sectors,
a
Digital
Citizen
Science
Observatory
(DiScO)
being
developed
by
Epidemiology
Population
Health
Laboratory
scaling
up
existing
infrastructure.
DiScO's
development
informed
Smart
Framework,
leverages
ethical
surveillance.
The
will
implementing
rapidly
adaptable,
replicable,
scalable
progressive
web
application
repurposes
jurisdiction-specific
cloud
infrastructure
jurisdictions.
designed
highly
adaptable
both
rapid
collection
as
well
responses
emerging
crises.
Data
sovereignty
decentralization
technology
core
aspects
observatory,
where
citizens
own
they
generate,
researchers
decision-makers
re-purpose
ultimate
aim
DiScO
transform
breaking
jurisdictional
silos
addressing
Язык: Английский