gqi.odfvmflines: Fibre Orientation Mapping Based on von Mises-Fisher Distributions with GQI Reconstruction

Description

In order to enable mapping complex white matter fibres in the brain, gqi.odfvmflines implements a new methodology based on directional statistics to estimate fibre profiles from high angular resolution diffusion imaging data. Statistical orientation estimation in gqi.odfvmf and gqi.odfvmflines is based on von Mises-Fisher clustering procedures provided by the R-package movMF, by Kurt Hornik and Bettina Gruen.

Usage

gqi.odfvmflines(gdi="gqi", fbase=NULL, roi=NULL,  rg=c(1,1), swap=FALSE,
 lambda=NULL, depth=3, btoption=2, threshold=0.4, kdir=6, zfactor=5,
 showglyph=FALSE, showimage="linesgfa", bview="coronal", bg="white",
 texture=NULL, clusterthr=0.6, aniso=NULL, ...)

Arguments

gdi

method of ODF reconstruction to use c("gqi", "gqi2") (default: "gqi").

fbase

Directory where the required input data files are located.

roi

Region of Interest (ROI) to use as mask; default mask (roi=NULL) uses an all brain mask for the supplied data set.

range of slices to process (default option rg=c(1,1)); rg=NULL processes all slices.

swap

toggle radiological/neurological orientation (default: FALSE).

lambda

diffusion sampling length in gdi="gqi" and gdi="gqi2". By default the following default values are used when lambda=NULL is specified: 1.25 in "gqi", 3 in "gqi2".

depth

sampling density on the hemisphere used in simulation (default N=321; depth=3).

btoption

b-table selection between btable.txt (btoption=1), and the 3D-DSI grid b-table extracted from the diffusion data set (data.bvec and data.bval). By default, the 3D-DSI grid b-table is used (btoption

threshold

thresholding generalized fractional anisotropy (GFA) value at each voxel (default: 0.4).

kdir

maximum number of fibre directions to map (default: 6).

zfactor

parameter controlling z-value in relief overlay maps (default: 5).

showglyph

logical variable controlling visualization of voxel glyphs (default: FALSE).

showimage

object controlling visualization of line-maps (default: "linesgfa"). Alternative options are: c("none", "gfa", "lines", "linesgfa", "linesrgbmap", "linesdata") (see Details).

bview

MRI slice view selection in {axial, coronal, sagittal} (default: "coronal").

map background colour (default "white").

texture

name of the PNG file to be used as RGB map overlay in some 'showimage' options (default NULL - no texture).

clusterthr

thresholding orientations based on ODF values at each voxel for directional clustering (default: 0.6).

aniso

anisotropic parameter in the range "[0,1)" or NULL to use in ODF pos-processing default: NULL.

...

additional material properties for geometry appearance as specified in rgl.material, or specification of non-default control parameters as detailed in movMF.

Value

gqi.odfvmflines produces line-maps of ODF profiles for diffusion data slices. The line-maps may be overlayed with generalized fractional anisotropy (GFA) relief maps, diffusion data maps or ROI maps. The file V1list.RData containing the first main orientation directions for all processed voxels is output for further posterior orientation processing.

concept

Diffusion Magnetic Resonance
GQI Reconstruction
von Mises Distributions
Fibre Orientation Mapping

Details

The function gqi.odfvmflines implements a mixture-model approach to clustering orientation distribution functions (ODFs) based on the von Mises-Fisher distributions. The method focus on clustering data on the unit sphere, where complexity arises from representing ODF profiles as directional data. GQI (Yeh et.al. 2010) or GQI2 (Garyfallidis 2012) may be used for ODF reconstruction.

Starting with the raw diffusion signal acquired on a grid of q-space, the ODF profile is estimated at each voxel, considering a sampling density of unit vectors on a unit S2 grid. When a threshold is applied to the estimated ODF at each voxel, the non-thresholded unit vectors provide directional statistics information about the estimated ODF profile. The main ODF orientations at each voxel relevant for fibre tracking may be estimated by clustering the non-thresholded unit vectors.

The main diffusion data set used in the examples is a DICOM data set provided by the "Advanced Biomedical MRI Lab, National Taiwan University Hospital", which is included in the "DSI Studio" package, publicly available from the NITRC repository (http://www.nitrc.org). Two b-tables defining the acquisition setup are specified. One is a b-table for a S2-like grid denoted by btable.txt. The other is the b-table for the 3D-DSI sampling scheme used in the DICOM data acquisition. This b-table has 203 points uniformly distributed on a 3D grid limited to the volume of the unit sphere. In both tables, the b-values range from 0 to 4000.

Slice map display and overlay selection is controlled by specifying one the arguments c("none", "gfa", "lines", "linesgfa", "linesrgbmap", "linesdata") for showimages. Meanings are as follows: "none" - no visualization; "gfa" - GFA map only; "lines" - line map only; "linesgfa" - GFA overlayed on line map; "linesrgbmap" - lines overlayed on RGB map (if available); "linesdata" - data_brain.nii.gz is overlayed on line map.

References

Ferreira da Silva, A. R., Modeling White-matter Fiber-orientation Uncertainty for Improved Probabilistic Tractography. In Proc. of the 2nd Int. Congr. on Neurotechnology, Electronics and Informatics, NEUROTECHNIX 2014, Rome, Italy, (Oct., 2014), A. Londral and P. Encarnacao, Eds., INSTICC, SCITEPRESS, pp 71-78.

Ferreira da Silva, A. R. Computational Representation of White Matter Fiber Orientations, International Journal of Biomedical Imaging, Vol. 2013, Article ID 232143, Hindawi Publishing Corporation http://dx.doi.org/10.1155/2013/232143.

Ferreira da Silva, A. R. Facing the Challenge of Estimating Human Brain White Matter Pathways. In Proc. of the 4th International Joint Conference on Computational Intelligence (Oct. 2012), K. Madani, J. Kacprzyk, and J. Filipe, Eds., SciTePress, pp. 709-714.

Hornik, K., and Gruen, B. movMF: Mixtures of von Mises-Fisher Distributions, 2014. R package version 0.2-0.

Yeh, F.-C., Wedeen, V. J., and Tseng, W.-Y. I. Generalized q-Sampling Imaging. IEEE Transactions on Medical Imaging 29, 9 (2010), 1626-1635.

Garyfallidis E., Towards an Accurate Brain Tractography, 2012, PhD Thesis, University of Cambridge.

Examples

Run this code

##-------------
  ## von Mises-Fisher fibre orientation mapping
  ## for a range of slices
  gqi.odfvmflines(gdi="gqi", rg=c(1,1), depth=2,
    showimage="linesdata", threshold=0.5)
  gqi.odfvmflines(gdi="gqi2", rg=c(1,1), depth=2,
    showimage="linesdata", threshold=0.5)
  ## display line-maps only
  gqi.odfvmflines(depth=2, showimage="lines")
  ## using GFA overlay
  gqi.odfvmflines(depth=2, showimage="linesgfa")
  ##-------------
  ## Show reconstructed glyphs in ODF processing 
  ## for principal direction determination
  gqi.odfvmflines(depth=3, showimage="linesdata",
    showglyph=TRUE, threshold=0.5)
  ## show glyphs with using 'aniso' parameter 
  gqi.odfvmflines(depth=3, showimage="linesdata",
    showglyph=TRUE, threshold=0.5, aniso=0.3)
  ##-------------
  ## using a ROI overlay
  gqi.odfvmflines(depth=3, roi="slfcst.nii.gz")
  ##-------------
  ## coronal view with texture for a single slice
  texturefname <- file.path(tempdir(),"rgbmap.png")
  rgbvolmap(texture=texturefname, bg="transparent")
  gqi.odfvmflines(threshold=0.5, showimage="linesrgbmap",
    texture=texturefname)
  ##-------------
  ## speeded up approximations: hardmax and common/numeric kappa
  gqi.odfvmflines(gdi="gqi", rg=c(1,1), depth=2,
    showimage="linesdata", threshold=0.5,
    E="hardmax", kappa=list(common = TRUE))
  gqi.odfvmflines(gdi="gqi", rg=c(1,1), depth=2,
    showimage="linesdata", threshold=0.5, E="hardmax", kappa=20)

Run the code above in your browser using DataLab