1. TRACULA Outputs
1.0.1. Remember...
For each new terminal that you open, you must do:
setenv SUBJECTS_DIR $TUTORIAL_DATA/diffusion_recons cd $TUTORIAL_DATA/diffusion_tutorial
1.1. Ouputs From FSL's dtifit
View the ouputs of the tensor fitting from FSL's dtifit, one of the Pre-processing steps. Some of these outputs, e.g., the FA or MD maps, can be used for voxel-based and ROI-based group analysis. They can be viewed/analyzed in either native space, CVS space, or MNI space. We'll take a look at all three.
1.1.1. Outputs In Native Diffusion Space
The diffusion tensor reconstruction files from FSL's dtifit in native space can be found in the dmri directory which was created when trac-all -prep was run. Let's cd into that directory for subject Diff001:
cd $TUTORIAL_DATA/diffusion_tutorial/Diff001/dmri
You can find all outputs from FSL's dtifit (tensor reconstruction) in this directory by using the ls command. The output maps are explained below:
- dtifit_FA.nii.gz - Fractional Anisotropy
- dtifit_MD.nii.gz - Mean Diffusivity
- dtifit_MO.nii.gz - Mode of the Anisotropy
- dtifit_S0.nii.gz - Non-Diffusion weighted image
- dtifit_L1.nii.gz - Primary Eigen Value
- dtifit_L2.nii.gz - Secondary Eigen Value
- dtifit_L3.nii.gz - Tertiary Eigen Value
- dtifit_V1.nii.gz - Primary Eigen Vector
- dtifit_V2.nii.gz - Secondary Eigen Vector
- dtifit_V3.nii.gz - Tertiary Eigen Vector
One could use the visualization tool, freeview, to view the Fractional Anisotropy image for Diff001 in native space. You can do this with the command below:
freeview $TUTORIAL_DATA/diffusion_tutorial/Diff001/dmri/dtifit_FA.nii.gz
Hover your mouse on different voxels throughout the image and notice the FA value change in the bottom toolbar. Look for areas of low FA and areas of high FA. Scroll through the slices using the Page Up and Page Down buttons. Feel free to change the contrast by holding down Shift and the left mouse button while dragging the mouse across the viewer window. When you are done, hit the X in the top right to close freeview.
All the above images are in the subject's native diffusion space. If you'd like to do a voxel-based group analysis of any of theses maps, they need to be transformed to a common space (CVS or MNI).
1.1.2. Outputs In CVS Space
The resampling of the dtifit output volumes to CVS space is performed as part of the tensor-fitting step in TRACULA preprocessing. This resampling using the transformation matrices created during the Intra- and Inter- registration steps of trac-all. It involves:
First, mapping the volumes from the individual's diffusion space to the individual's anatomical space. This uses the affine transform that was computed by the Intra-subject registration Pre-processing step in TRACULA. This transform can be computed using either FSL's FLIRT or the in-house registration algorithm bbregister. (The latter can be performed only if you have a surface reconstruction from FreeSurfer.)
Second, mapping the volumes from the individual's anatomical space to the space of the CVS template (cvs_avg35 by default). This uses the non-linear transform computed during the Inter-subject registration Pre-processing step using mri_cvs_register.
Among dtifit outputs, only the scalar maps, i.e., all outputs except the eigenvectors, are transformed into CVS space. The transformed volumes can be found in the following directory which you can navigate to:
cd $TUTORIAL_DATA/diffusion_tutorial/Diff001/dmri/cvs/
It is always a good idea to check how well the registration is between each subject and the template. A bad registration could mean inaccurate tractography results. Since we ran both bbregister and FLIRT for Intra-subject registration of the FA, you could look at the registration between the FA image and the CVS template for either or both.
With this command, you can view the registration between the FA and the CVS template for the bbregister-ed FA version:
freeview -v $FREESURFER_HOME/subjects/cvs_avg35/mri/norm.mgz \ $TUTORIAL_DATA/diffusion_tutorial/Diff001/dmri/cvs/dtifit_FA.bbr.nii.gz
Or if you prefer to check the registration between the FA image and CVS template using the FLIRT registered FA, try this command:
freeview -v $FREESURFER_HOME/subjects/cvs_avg35/mri/norm.mgz \ $TUTORIAL_DATA/diffusion_tutorial/Diff001/dmri/cvs/dtifit_FA.flt.nii.gz
To see the correspondence between the FA in CVS space and the CVS template and check the accuracy of the registration, toggle between the two images by using the 'Alt + C' hot key. Or you can check and uncheck the box next to volume at the top of the list on the left panel.
Alternatively, you can adjust the opacity:
You can make one of the views bigger by choosing the 1x1 button (). Use the orientation buttons to switch the plane (
,
,
,
). Put your cursor at the edge of a gyrus or ventricle as a landmark before toggling back and forth between the two volumes. Scroll through the slices and try it again on a different section of the brain. Notice how accurate the registration is - the anatomy in individual subject's image lines up well with the anatomy in the template's image. When you are done, be sure to close freeview again.
1.1.3. Outputs In MNI space
The resampling of the dtifit output volumes to MNI space is performed as part of the tensor-fitting step in TRACULA preprocessing. This is an alternative template space to the CVS space. This resampling involves:
- First, mapping the volumes from the individual's diffusion space to the individual's anatomical space, as described above for CVS, using either bbregister, FLIRT, or both.
- Second, mapping the volumes from the individual's anatomical space to the space of the MNI template (MNI152_T1_1mm_brain by default). This uses the affine transform that is computed by the Inter-subject registration step in TRACULA preprocessing, using FSL's FLIRT.
Again, among dtifit outputs, only the scalar maps, i.e., all outputs except the eigenvectors, are transformed to MNI space. The transformed volumes can be found in the following directory.
You can find the MNI registration results in the following directory:
cd $TUTORIAL_DATA/diffusion_tutorial/Diff001/dmri/mni/
- To check the registration of the FA image to the MNI template in freeview, as you did for CVS, use the commands below.
View the registration between the FA and the MNI template for the bbregister-ed FA version:
freeview -v $FSLDIR/data/standard/MNI152_T1_1mm_brain.nii.gz \ $TUTORIAL_DATA/diffusion_tutorial/Diff001/dmri/mni/dtifit_FA.bbr.nii.gz
Or check the registration between the FLIRT-registered FA image and the MNI template:
freeview -v $FSLDIR/data/standard/MNI152_T1_1mm_brain.nii.gz \ $TUTORIAL_DATA/diffusion_tutorial/Diff001/dmri/mni/dtifit_FA.flt.nii.gz
Again, use the cursor as a landmark and toggle (Alt+C) between the FA and the template to verify the accuracy of the registration. Be sure to close the Freeview window when you are finished viewing the data before proceeding with the next step.
1.2. FSL's bedpostX Outputs
For a detailed overview of the outputs given by FSL's bedpostX please refer to the link below,
http://www.fmrib.ox.ac.uk/fsl/fdt/fdt_bedpostx.html
1.3. Tractography Outputs
1.3.1. Visualizing The Posterior Distribution Of A White Matter Pathway
To visualize the posterior distribution of the inferior logitudinal fasiculus tract (ILF) that was estimated using training data in MNI space, do the following:
freeview -v $TUTORIAL_DATA/diffusion_tutorial/Diff001/dmri/dtifit_FA.nii.gz \ $TUTORIAL_DATA/diffusion_tutorial/Diff001/dpath/lh.ilf_AS_avg32_mni_flt/path.pd.nii.gz:colormap=jet:isosurface=0,0:color='yellow' \ $TUTORIAL_DATA/diffusion_tutorial/Diff001/dpath/rh.ilf_AS_avg32_mni_flt/path.pd.nii.gz:colormap=jet:isosurface=0,0:color='yellow'
Click on the 3D button ()to view the tract. Use the 1x1 button (
) to make it fill the window. You can rotate the image by grabbing it with the mouse and dragging. You can also move the slice planes by grabbing those and dragging them out of the way. Hold down the middle button and push the mouse forward for a smooth zoom inwards.
You can play with the isosurface threshold to change the isosurface.
You can also adjust the heatmap thresholds for the posterior distribution of ILF. You'll want to go back to the non-3D views (,
,
) to see the effect.
DO NOT DO THE BELOW
1.3.2. Visualizing The Posterior Distribution Of All Reconstructed Tracts
Tracula also outputs a merged 4D volume, which is a concatenation of the estimated distribution of all the tracts. To visualize all tracts reconstructed using TRACULA, do the following:
freeview -tv $TUTORIAL_DATA/diffusion_tutorial/Diff001/dpath/merged_avg32_mni_flt.mgz \ -v $TUTORIAL_DATA/diffusion_tutorial/Diff001/dmri/dtifit_FA.nii.gz
The lower-left corner of the freeview display gives the color lookup table for all the tracts. The tracts in the merged volume are displayed at 20% of their maximum threshold, by default. Hence the posterior distribution of some of tracts may not be displayed in its entirety. To look at the complete path distribution of a specific tract in non-3D views, you can play with the threshold values again:
The threshold for each tract can also be adjusted in the 3D view by selecting a specific tract from the lookup table and adjusting the threshold button
at the bottom of the freeview diplay panel to view the 3D distribution of a tract at various thresholds (between 1 and 100).
Now that you've seen what the output looks like, you can take a look at the resulting stats for each tract by clicking on the Next button below.