Electroencephalography Estimates Brain Age in Infants with High Precision: Leveraging Advanced Machine Learning in Healthcare
NeuroImage,
Journal Year:
2025,
Volume and Issue:
unknown, P. 121200 - 121200
Published: April 1, 2025
Changes
in
the
pace
of
neurodevelopment
are
key
indicators
atypical
maturation
during
early
life.
Unfortunately,
reliable
prognostic
tools
rely
on
assessments
cognitive
and
behavioral
skills
that
develop
towards
second
year
life
after.
Early
assessment
brain
using
electroencephalography
(EEG)
is
crucial
for
clinical
intervention
care
planning.
We
developed
a
methodology
conventional
machine
learning
(ML)
novel
deep
(DL)
networks
to
efficiently
quantify
difference
between
chronological
biological
age,
so-called
age
gap
(BAG)
as
marker
accelerated/decelerated
development.
In
this
cross-sectional
study,
EEG
from
219
typically-developing
infants
aged
three
14-months
was
used.
For
DL
networks,
input
samples
were
increased
2,628
recordings.
further
validated
BAG
tool
population
at
risk
with
abnormal
growth
(macrocephaly)
capture
deviation
normal
aging.
Our
results
indicate
outperform
ML
models,
capturing
complex
non-monotonic
characteristics
predicting
mean
absolute
error
only
one
month
(MAE=1
month,
95%CI:0.88-1.15,
r=0.82,
95%CI:0.78-0.85).
Additionally,
developing
follows
trajectory
characterized
by
non-linearity
complexity
which
alpha
rhythm
plays
an
important
role.
could
detect
group-level
maturational
delays
macrocephaly
(pvalue=0.009).
macrocephaly,
negatively
correlated
general
adaptive
composite
ABAS-II
(pvalue=0.04)
18-months
information
processing
speed
scale
WPSSI-IV
four
(pvalue=0.006).
The
EEG-based
score
offers
non-invasive
measure
maturation,
significant
advantages
implications
developmental
neuroscience
practice.
Language: Английский
Resting state aperiodic slope predicts Research Domain Criteria Social Processes in young typically developing adults
bioRxiv (Cold Spring Harbor Laboratory),
Journal Year:
2023,
Volume and Issue:
unknown
Published: Sept. 22, 2023
Abstract
Background
The
aperiodic
exponent
has
been
utilised
to
demonstrate
differences
in
brain
excitation-inhibition
(E/I)
ageing,
cognition,
and
neuro-
psycho-pathology.
Furthermore,
E/I
imbalance
associated
with
social
communication
difficulties
clinical
non-clinical
cohorts.
No
work
date,
however,
explored
the
association
between
activity
difficulties.
Methods
A
total
of
40
typically
developing
young
adults
aged
18-24
(24
female;
age
mean=21.6,
SD=1.9)
each
underwent
5
minutes
eyes
open
closed
resting
state
EEG
using
a
64-channel
HydroCel
Geodesic
Sensor
Net.
Participants
also
completed
Research
Domain
Criteria
Social
Processes
measures:
Responsiveness
Scale
(SRS),
Reading
Mind
Eyes
task
(RME),
Penn
Emotional
Recognition
Task
(ER-40).
Results
Sex
offset
were
observed,
steeper
greater
observed
males,
suggestive
inhibitory
tone
(
p
s<.05).
Whole-brain
was
not
SRS,
RME,
or
ER-40.
There
some
support
for
increased
SRS
scores
right
(suggestive
excitatory
tone),
this
significant
=.068).
Conclusions
These
preliminary
data
show
promise
supporting
notion
that
is
processing
difficulties,
larger
samples
are
needed
validate
these
findings.
Language: Английский