Adaptive smoothing of DWI data. Smoothing is performed both in space and on the sphere (e.g. between images obtained for different gradient directions) employing a natural
geometrical distance ( in SE(3)). Structural adaptation is used in space only.
Method dwi.smooth
refers to the original POAS approach for single shell data.
Method dwi.smooth.ms
implements an improved method that is applicable for both single and multi-shell data.
# S4 method for dtiData
dwi.smooth(object, kstar, lambda=20, kappa0=NULL, mask=NULL, ncoils=1,
sigma=NULL, level=NULL, vred=4, verbose=FALSE, dist=1,
model=c("Gapprox","Gapprox2","Chi","Chi2"))
# S4 method for dtiData
dwi.smooth.ms(object, kstar, lambda=12, kappa0=.5, ncoils=1,
sigma=NULL, ws0=1, level=NULL, mask = NULL, xind=NULL,
yind=NULL, zind=NULL, verbose=FALSE,
usemaxni=TRUE, memrelease = TRUE)
An object of class "dtiData"
with smoothed diffusion weighted images.
Object of class "dtiData"
Number of steps in structural adaptation
Scale parameter in adaptation
determines amount of smoothing on the sphere. Larger values correspond to stronger smoothing on the sphere. If kappa0=NULL
a value is that corresponds to a variace reduction with factor vred
on the sphere.
Number of coils in MR system
Error standard deviation. Assumed to be known and homogeneous in the current implementation. A reasonable estimate may be defined
as the modal value of standard deviations obtained using method getsdofsb
.
Threshold for image intensities when setting mask.
Binary 3D image defining a mask
Used if kappa0=NULL
to specify the variance reduction on the sphere when suggesting a value of kappa0
index for x-coordinate
index for y-coordinate
index for z-coordinate
If verbose=TRUE
additional reports are given.
Distance in SE3. Reasonable values are 1 (default, see Becker et.al. 2012), 2 ( a slight modification of 1: with k6^2 instead of abs(k6)) and 3 (using a 'naive' distance on the sphere)
Determines which quantities are smoothed. Possible values are
"Chi"
for observed values (assumed to be distributed as noncentral
Chi with 2*ncoils
degrees of freedom), "Chi2"
for squares of
observed values (assumed to be distributed as noncentral
Chi-squared with 2*ncoils
degrees of freedom). "Gapprox"
and "Gapprox2"
use a Gaussian approximation for the noncentral
Chi distribution to smooth ovserved and squared values, respectively.
Factor to downweight information from S0 images, defaults
to 1/numer of s0 images
.
If "usemaxni==TRUE"
a strikter penalization is used.
If "memrelease==TRUE"
try to release allocated memory whenever possible.
signature(object) = "ANY"
Returns a warning.
signature(object) = "dtiData"
Smoothing of DWI data
Karsten Tabelow tabelow@wias-berlin.de
J\"org Polzehl polzehl@wias-berlin.de
dtiData
,
dtiData
,