Differences between revisions 19 and 20
Deletions are marked like this. Additions are marked like this.
Line 19: Line 19:
{{{
AnatomiCuts -s segmentation.nii.gz -f streamlines.vtk -labels -o outputFolder <options>
}}}
Where
Line 20: Line 24:
-s segmentation file such as wm2009parc.nii.gz
Line 21: Line 26:
-f streamline file in trk format.

-c number of clusters

-s segmentation file such as wm2009parc.nii.gz

-f streamline file in trk format.

-c number of clusters. Default is 200.

-n number of points to be used per streamline. Default is 10.

-e number of streamlines to be used for the eigendecomposition of normalized cuts algorith. Default is 500. Using less number of streamlines will fasten substantially the algorithm at the expense of accuracy. For a quick test 50 could be ok, but in practice less than 300 is not recommended (See Siless et al, 2018).

-d directional neighbors to be used, default is "a" all (26), diagonal "d" is 14 and straight "s" is 6. (See Siless et al 2018).
sters. Default is 200.

-n number of points to be used per streamline. Default is 10.

-e number of streamlines to be used for the eigendecomposition of normalized cuts algorith. Default is 500. Using less number of streamlines will fasten substantially the algorithm at the expense of accuracy. For a quick test 50 could be ok, but in practice less than 300 is not recommended (See Siless et al, 2018).

-d directional neighbors to be used, default is "a" all (26), diagonal "d" is 14 and straight "s" is 6. (See Siless et al 2018).

AnatomiCuts

AnatomiCuts is an unsupervised hierarchical clustering that uses an anatomical similarity metric.

  • anatomiCuts_tree.jpgrainbow_brain.png

Inputs

- streamllines file (*.trk) - a segmentation image, preferable including cortical and subcortical parcellation with white matter segmentation based on neighboring regions (wmparc.mgz or wm2009parc.mgz).

It is important that both files are in the same space. To validate this you can open the streamline file and the segmentation using both Freeview and Trackvis, since Freeview does not take into account RAS orientation, and Trackvis does not take into account the image offset.

Filtering streamlines

Sometimes deterministic tractography can generate spurious streamlines outside the brain. To exclude them we can use streamlineFilter. It is recommended to remove the short streamlines that can prematurely end during tractography. It is also possible to remove the ushape streamlines.

Running AnatomiCuts

AnatomiCuts -s segmentation.nii.gz -f streamlines.vtk  -labels  -o outputFolder  <options> 

Where

-s segmentation file such as wm2009parc.nii.gz

-f streamline file in trk format.

-c number of clusters

-s segmentation file such as wm2009parc.nii.gz

-f streamline file in trk format.

-c number of clusters. Default is 200.

-n number of points to be used per streamline. Default is 10.

-e number of streamlines to be used for the eigendecomposition of normalized cuts algorith. Default is 500. Using less number of streamlines will fasten substantially the algorithm at the expense of accuracy. For a quick test 50 could be ok, but in practice less than 300 is not recommended (See Siless et al, 2018).

-d directional neighbors to be used, default is "a" all (26), diagonal "d" is 14 and straight "s" is 6. (See Siless et al 2018). sters. Default is 200.

-n number of points to be used per streamline. Default is 10.

-e number of streamlines to be used for the eigendecomposition of normalized cuts algorith. Default is 500. Using less number of streamlines will fasten substantially the algorithm at the expense of accuracy. For a quick test 50 could be ok, but in practice less than 300 is not recommended (See Siless et al, 2018).

-d directional neighbors to be used, default is "a" all (26), diagonal "d" is 14 and straight "s" is 6. (See Siless et al 2018).

References

AnatomiCuts: Hierarchical clustering of tractography streamlines based on anatomical similarity. Siless V., Chang K., Fischl B., Yendiki A.. NeuroImage 2018.

AnatomiCuts (last edited 2019-06-05 04:35:50 by VivianaSiless)