Hi,

It is fine to use this file.
Just take note of the value (intensity) in the voxel of interest and then look in the $FSLDIR/data/atlases/JHU-tracts.xml
for a line where the index value is *one less than* the number you see in FSLView.  That is, if you see the number 4 in FSLView then you look for the index 3 in this file, since in the xml file they start counting the indices from 0 rather than 1.

All the best,
        Mark

On 23 Jan 2014, at 20:23, Bedda Rosario <[log in to unmask]> wrote:

> Hello,
>
> I have run several TBSS analysis on FA data, and I have used randomise option to perform two sample t-tests.  Now, I would like to determine which tracts show a statistically significant difference between the groups.
>
> For this, I have created a mask image using the tbss_tfce_corrp_tstat result  image, everything that is above 0.95 there is a 1.  Then, I multiplied this mask by the JHU-ICBM-T2-1mm.nii.gz atlas to get the list of tracts.
>
> However, when I checked the results there are other regions/tracts that shows statistically significant differences between the groups but are not part of this atlas.
>
> Is it okay to use the atlas JHU-ICBM-tracts-maxprob-thr0-1mm to find the labels that are not included in the JHU-ICBM-T2-1mm.nii.gz atlas?
>
> If so, where I can find the list of white matter tract labels?
>
> I would appreciate any help.
>
> Thanks,
> Bedda