bioRxiv (Cold Spring Harbor Laboratory),
Journal Year:
2023,
Volume and Issue:
unknown
Published: May 14, 2023
Abstract
One
fundamental
challenge
in
diffusion
MRI
(dMRI)
harmonization
is
to
disentangle
the
contributions
of
scanner-related
effects
from
variable
brain
anatomy
for
observed
imaging
signals.
Conventional
methods
rely
on
establishing
an
atlas
space
resolve
anatomical
variability
and
generate
a
unified
inter-site
mapping
function.
However,
this
approach
limited
accounting
misalignment
neuroanatomy
that
still
widely
persists
even
after
registration,
especially
regions
close
cortical
boundaries.
To
overcome
challenge,
we
propose
personalized
framework
paper
more
effectively
address
confounding
dMRI
harmonization.
Instead
using
common
template
representing
site-effects
all
subjects,
main
novelty
our
method
adaptive
computation
templates
both
source
target
scanning
sites
estimate
We
integrate
with
rotation
invariant
spherical
harmonics
(RISH)
features
achieve
signals
between
sites.
In
experiments,
proposed
applied
harmonize
data
acquired
two
platforms:
Siemens
Prisma
GE
MR750
Adolescent
Brain
Cognitive
Development
(ABCD)
dataset
compared
state-of-the-art
based
RISH
features.
Our
results
indicate
achieves
superior
performance
not
only
reducing
variations
due
scanner
differences
but
also
preserving
sex-related
biological
original
cohorts.
Moreover,
assess
impact
estimation
fiber
orientation
distributions
(FOD)
show
robustness
procedure
bioRxiv (Cold Spring Harbor Laboratory),
Journal Year:
2024,
Volume and Issue:
unknown
Published: Aug. 19, 2024
Adolescent
neuroimaging
studies
of
sex
differences
in
the
human
brain
predominantly
examine
mean
between
males
and
females.
This
focus
on
between-groups
without
probing
relative
distributions
similarities
may
contribute
to
both
conflation
overestimation
sexual
dimorphism
developing
brain.
We
aimed
characterize
variance
macro-
micro-structure
early
adolescence
as
it
pertains
at
birth
using
a
large
sample
9-11
year-olds
from
Brain
Cognitive
Development
(ABCD)
Study
(N=7,723).
Specifically,
for
global
regional
estimates
gray
white
matter
volume,
cortical
thickness,
microstructure
(i.e.,
fractional
anisotropy
diffusivity),
we
examined:
within-
between-sex
variance,
overlap
male
female
distributions,
inhomogeneity
via
Fligner-Killeen
test,
an
analysis
(ANOSIM).
For
completeness,
examined
these
uncorrected
(raw)
residualized
after
mixed-effects
modeling
account
age,
pubertal
development,
socioeconomic
status,
race,
ethnicity,
MRI
scanner
manufacturer,
total
where
applicable.
The
was
universally
greater
than
difference
(overlap
coefficient
range:
0.585
-
0.985)
ratio
within-sex
similar
(ANOSIM
R
-0.001
0.117).
All
subcortical
volumes
showed
significant
whereas
minority
regions
anisotropy,
diffusivity.
Inhomogeneity
reduced
accounting
other
sources
variance.
Overlap
coefficients
were
larger
ANOSIM
values
smaller
outcomes,
indicating
once
covariates.
Reported
adolescent
structure
be
driven
by
disparities
rather
binary,
sex-based
phenotypes.
Contrary
popular
view
sexually
dimorphic,
found
more
similarity
sexes
all
measurements
examined.
study
builds
upon
previous
findings
illustrating
importance
considering
when
examining
structure.
Research Square (Research Square),
Journal Year:
2024,
Volume and Issue:
unknown
Published: Oct. 17, 2024
Abstract
Background
Adolescent
neuroimaging
studies
of
sex
differences
in
the
human
brain
predominantly
examine
mean
between
males
and
females.
This
focus
on
between-groups
without
probing
relative
distributions
similarities
may
contribute
to
both
conflation
overestimation
sexual
dimorphism
developing
brain.
Methods
We
aimed
characterize
variance
macro-
micro-structure
early
adolescence
as
it
pertains
at
birth
using
a
large
sample
9-11
year-olds
from
Brain
Cognitive
Development
(ABCD)
Study
(N=7,723).
Specifically,
for
global
regional
estimates
gray
white
matter
volume,
cortical
thickness,
microstructure
(i.e.,
fractional
anisotropy
diffusivity),
we
examined:
within-
between-sex
variance,
overlap
male
female
distributions,
inhomogeneity
via
Fligner-Killeen
test,
an
analysis
(ANOSIM).
For
completeness,
examined
these
uncorrected
(raw)
residualized
after
mixed-effects
modeling
account
age,
pubertal
development,
socioeconomic
status,
race,
ethnicity,
MRI
scanner
manufacturer,
total
where
applicable.
Results
The
was
universally
greater
than
difference
(overlap
coefficient
range:
0.585
-
0.985)
ratio
within-sex
similar
(ANOSIM
R
-0.001
0.117).
All
subcortical
volumes
showed
significant
whereas
minority
regions
anisotropy,
diffusivity.
Inhomogeneity
reduced
accounting
other
sources
variance.
Overlap
coefficients
were
larger
ANOSIM
values
smaller
outcomes,
indicating
once
covariates.
Conclusions
Reported
adolescent
structure
be
driven
by
disparities
rather
binary,
sex-based
phenotypes.
Contrary
popular
view
sexually
dimorphic,
found
more
similarity
sexes
all
measurements
examined.
study
builds
upon
previous
findings
illustrating
importance
considering
when
examining
structure.
Typical
sex
differences
in
white
matter
(WM)
microstructure
during
development
are
incompletely
understood.
Here
we
evaluated
WM
typical
brain
using
a
sample
of
neurotypical
individuals
across
wide
developmental
age
(N=239,
aged
5-22
years).
We
used
the
conventional
diffusion-weighted
MRI
(dMRI)
model,
diffusion
tensor
imaging
(DTI),
and
two
advanced
dMRI
models,
distribution
function
(TDF)
neurite
orientation
dispersion
density
(NODDI)
to
assess
microstructure.
exhibited
significant,
regionally
consistent
development.
Additionally,
TDF
model
was
most
sensitive
detecting
differences.
These
findings
highlight
importance
considering
neurodevelopmental
research
underscore
value
model.
Human Brain Mapping,
Journal Year:
2023,
Volume and Issue:
44(10), P. 3986 - 3997
Published: May 1, 2023
Abstract
This
study
investigated
associations
between
psychological
resilience
and
characteristics
of
white
matter
microstructure
in
pediatric
concussion.
is
a
case
control
planned
substudy
larger
randomized
controlled
trial.
Children
with
an
acute
concussion
or
orthopedic
injury
were
recruited
from
the
emergency
department.
Participants
completed
both
Connor–Davidson
Resilience
Scale
10
MRI
at
72
h
4‐weeks
post‐injury.
The
association
resiliency
fractional
anisotropy
(FA),
mean
diffusivity
(MD),
axial
(AD),
radial
(RD)
timepoints
examined.
We
examined
whether
these
moderated
by
group.
captured
diffusion
tensor
imaging
metrics
4
weeks
was
also
investigated.
Clusters
extracted
using
significance
threshold
threshold‐free
cluster
enhancement
corrected
p
<
.05.
A
total
66
children
(median
(IQR)
age
=
12.88
(IQR:
11.80–14.36);
47%
female)
29
orthopedic‐injury
12.49
11.18–14.01);
41%
included.
negative
correlation
identified
group
FA.
Meanwhile,
positive
correlations
MD
RD.
These
findings
suggest
that
associated
forceps
minor,
superior
longitudinal
fasciculus,
anterior
thalamic
radiation
post‐concussion.
seems
to
be
neural
integrity
only
phase
thus
may
considered
when
researching
recovery.
bioRxiv (Cold Spring Harbor Laboratory),
Journal Year:
2023,
Volume and Issue:
unknown
Published: May 14, 2023
Abstract
One
fundamental
challenge
in
diffusion
MRI
(dMRI)
harmonization
is
to
disentangle
the
contributions
of
scanner-related
effects
from
variable
brain
anatomy
for
observed
imaging
signals.
Conventional
methods
rely
on
establishing
an
atlas
space
resolve
anatomical
variability
and
generate
a
unified
inter-site
mapping
function.
However,
this
approach
limited
accounting
misalignment
neuroanatomy
that
still
widely
persists
even
after
registration,
especially
regions
close
cortical
boundaries.
To
overcome
challenge,
we
propose
personalized
framework
paper
more
effectively
address
confounding
dMRI
harmonization.
Instead
using
common
template
representing
site-effects
all
subjects,
main
novelty
our
method
adaptive
computation
templates
both
source
target
scanning
sites
estimate
We
integrate
with
rotation
invariant
spherical
harmonics
(RISH)
features
achieve
signals
between
sites.
In
experiments,
proposed
applied
harmonize
data
acquired
two
platforms:
Siemens
Prisma
GE
MR750
Adolescent
Brain
Cognitive
Development
(ABCD)
dataset
compared
state-of-the-art
based
RISH
features.
Our
results
indicate
achieves
superior
performance
not
only
reducing
variations
due
scanner
differences
but
also
preserving
sex-related
biological
original
cohorts.
Moreover,
assess
impact
estimation
fiber
orientation
distributions
(FOD)
show
robustness
procedure