dwiMixtensor-methods: Methods for Function `dwiMixtensor' in Package `dti'

Description

The method estimates, in each voxel, a mixture of radial symmetric tensors from the DWI data contained in an object of class "dtiData".

Usage

## S3 method for class 'dtiData':
dwiMixtensor(object, maxcomp=3,  p=40, method="mixtensor", reltol=1e-6,
          maxit=5000, ngc=1000, optmethod="BFGS",nguess=100*maxcomp^2,
          msc="BIC",pen=NULL)
## S3 method for class 'dwiMixtensor,dtiData':
dwiMtImprove(mtobj, dwiobj, maxcomp=3,  p=40, method="mixtensor", reltol=1e-6, maxit=5000,ngc=1000, optmethod="BFGS", nguess=100*maxcomp^2,msc="BIC",pen=1,where=NULL)
## S3 method for class 'dwiMixtensor,dwiMixtensor':
dwiMtCombine(mtobj1,mtobj2, msc="BIC", where=NULL)

Arguments

object

Object of class "dtiData"

maxcomp

Maximal number of mixture components.

Exponent in Jian-Model (only effective if method="Jian"(Not yet implemented)

method

Specifies the mixture model used. method="mixtensor" specifies a mixture of tensor models (default), method="Jian" refers to the model defined in Jian et al. (2007) with fixed p. For method dwiMtImprove

reltol

Relative tolerance for R's nmmin() function.

maxit

Maximal number of iterations in Rs nmmin() function.

ngc

provide information on number of voxel processed, elapsed time and estimated remaining time after ngc voxel.

optmethod

Optimization method used, currently available is optmethod="Nelder-Mead", optmethod="BFGS" using analytic gradients will be added

nguess

number of guesses in search for initial estimates

msc

Criterion used to select the order of the mixture model, either BIC (Bayes Information Criterion) AIC (Akaike Information Criterion) or AICC ((Bias-)Corrected Akaike Information Criterion)

pen

Penalty used in optimization criterion for negative mixture components.

mtobj

For method "dwiMtImprove" an initial "dwiMixtensor"-object.

dwiobj

For method "dwiMtImprove" the "dwiData" object corresponding to mtobj

mtobj1

For method "dwiMtCombine" an "dwiMixtensor"-object.

where

Mask of voxel for which "dwiMtImprove" or "dwiMtCombine" should be performed.

mtobj2

For method "dwiMtCombine" an "dwiMixtensor"-object obtained from the same "dwiData" object. The maximum number of components in mtobj2 should preferably be less or equal to the maximum number of compone

Value

An object of class "dwiMixtensor".

Details

The method "dwiMixtensor" estimates, in each voxel, a mixture of radial symmetric tensors from the DWI data contained in an object of class "dtiData". The number of mixture components is selected depending on the data, with a maximum number of components specified by maxcomp.

If method="Jian" the model parameters from Jian et al. (2007) with fixed p are estimated. With method="Jian2" also p is estimated.

In a voxel tensors are restricted to be rotational sysmmetric with common excentricity and destinct largest eigenvalue. The method "dwiMtImprove" evaluates the results in mtobj, including directions identified in neighboring voxel, to obtain an alternative set of initial values for estimating the parameters of the tensor mixture model. The resulting object contains, in a voxel, either the results from mtobj or the estimated parameters obtained by optimisation starting with the new initial values, depending on a comparison of their respective estimated MSEP values as specified by msc. The specification of method="mixtensoriso" includes an isotropic component into the model. If the isotropic term leads to an improbement with respect to the estimated MSEP the sum of weights for such a voxel will be less than 1, with the discrepancy to 1 corresponding to the partial volume associated to the isotropic compartment. The method "dwiMtCombine" enables to combine results obtained for the same dwi data set with different specifications, e.g. for maximum number of components mcomp and settings that influence initial estimates. The combined result contains in each voxel the best result from both reconstructions with respect to the specified model selection criterion msc.

References

Jian et al. (2007), A novel tensor distribution model for the diffusion-weighted MR signal, NeuroImage 37, 164--176.

Examples

Run this code

demo(mixtens_art)

Run the code above in your browser using DataLab