Print

Print
Hi Rosalia,

I actually haven't tried the function "atlasquery" before, but now I find it quite handy.

Here's an example:

$ atlasquery -a "Harvard-Oxford Subcortical Structural Atlas" -m MNI152_T1_2mm-mask.nii.gz 
Left Cerebral White Matter:20.5
Left Cerebral Cortex :44

The mask file here includes only two voxels: (GM 39%; WM 22%) and (GM 49%, WM 19%) in HO-sub atlas, so it shows the averages of probability labels for a given mask.

Now you need to make a binary mask with suprathreshold (say, alpha=0.05) voxels:

$ fslmaths <1-pval_result> -uthr 0.95 -bin <mask>

and put this <mask> file into atlasquery:

$ atlasquery -a "JHU White-Matter Tractography Atlas" -m <mask>

and you get a list of averaged probabilities.

Best,
---
Seung-Goo KIM

On 3, Apr, 2013, at 10:50 AM, Rosalia Dacosta Aguayo wrote:

Hi Seung-Goo

Many thanks for your kind and helping explantion. If you do not mind, I am going to send you two images to your personal mail because FSL mailing list do will not support them....with the coordinates...in order to guess how can I get more information about the cluster.

Best regards,
Rosalia.

2013/4/2 SG KIM <[log in to unmask]>
Hi Rosalia,

I guess you actually told us the voxel indices of (55,125,38), which corresponds to (35.0, -2.0, -34.0) MNI-coordinate and according to other atlas tool values: white matter voxel in the anterior part of right temporal lobe.

If it is correct, first you should know that neither ICBM-DTI-81 nor JHU WM atlas cover whole white matter voxels, because the atlases only specify some of well-known major fiber bundles; thus many of white matter voxels in fine tracts are not classified for many issues.

And second, you may want to look at the values from other voxels of the cluster you found, not only the one peak voxel. You may have more information about the cluster (say, whether it is likely to be part of uncinate fasciculus or inferior longitudinal fasiculus).


Best,
---
Seung-Goo KIM

On 2, Apr, 2013, at 10:13 PM, Rosalia Dacosta Aguayo wrote:


Dear FSL experts, sorry for posting this mail again.

I am having some problems with the anatomical identification of some of my tbss results.

For example: one of my results give me the following MNI coordinates: 55-124-38 of a cluster size of 1758 mm(3).

When I open Fslview, load the image and write the coordinates, the message I get is "label not found":

Harvard-Oxford Subcortical Structural Atlas
69% Right Cerebral White Matter, 30% Right Cerebral Cortex
JHU ICBM-DTI-81 White-Matter Labels
Unclassified
JHU White-Matter Tractography Atlas
No label found

Any one could help me with this?

Yours sincerely,
Rosalia.