Hi Rosalia,

RD isn't produced by dtifit, so if you calculate it by hand (which is fine), you need to make sure that these files are named exactly in the same way the other dtifit outputs (L1, L2, L3, MD, V1, etc). Check the other filenames you have in your TBSS directory and try to emulate the exact same names. It should work just fine.

All the best,
Anderson

2013/1/14 Rosalia Dacosta Aguayo <[log in to unmask]>

Hi Anderson,

I run the four steps for calculating FA, but not for calculating the other indexes. I used FA index to calculate the other indexes with tbss_non_FA RD...for example (first I calculated RD = L2 + L3) / 2 and then tbss_non_FA RD....for other indexes different of FA some experts told me I had not to run the four steps again.....

I am very confused now....

Yours sincerely,

Rosalia.

2013/1/14 Anderson M. Winkler <[log in to unmask]>

Hi Rosalia,

You need to run all the steps before trying to do non-FA analyses. Please, see the instructions here: http://fsl.fmrib.ox.ac.uk/fsl/fslwiki/TBSS/UserGuide#Using_non-FA_Images_in_TBSS

About GLM: randomise uses the general linear model for the regression, just as fsl_glm, FEAT and many other applications do. What randomise does differently is not in the regression stage, but later, when the p-values are computed. So, when using randomise, yes, the GLM is being used behind the scenes too.

All the best,

Anderson

2013/1/14 Rosalia Dacosta Aguayo <[log in to unmask]>

Sorry Tibor,

Thanks for your answer.

But when I calculated the other indexes, I did not get all_RD_skelotonized file...because I did not run  the four TBSS steps, only tbss_non_FA RD...for example. So, where is this file??

Yes, you are right, I mentioned randomise thinking about the Glm design (design.con and design.mat).

Best regards,

Rosalia.

2013/1/14 Tibor Auer <[log in to unmask]>

Only all_FA_skeletonised must be replaced (e.g. all_RD_skeletonised), mean_FA_skeleton_mask remains.

 

One remark: if I am correct, randomise is not GLM.

 

Cheers,

 

Auer, Tibor M.D. Ph.D.

MRC Cognition and Brain Sciences Unit
15 Chaucer Road
Cambridge
CB2 7EF

United Kingdom

Phone/Work: +44-(0)1223-355294 (ext. 869)

Mobile: +44-(0)7906-863837

Mail: [log in to unmask]

 

From: FSL - FMRIB's Software Library [mailto:[log in to unmask]] On Behalf Of Rosalia Dacosta Aguayo
Sent: Monday, January 14, 2013 6:14 PM
To: [log in to unmask]
Subject: [FSL] One doubt running randomise with other indexes than FA

 

Dear Fsl experts,

My apologizes for this basic question.

If I want to run randomise with other indexes than FA, it would be right running the following command:

 randomise -i all_FA_skeletonised -o tbss -m mean_FA_skeleton_mask -d design.mat -t design.con -n 5000 --T2 -V

Entering in the Glm, instead of FA,  the RD or MD, AD?

Many thanks in advance for your helping.

Your sincerely,

Rosalia Dacosta.