Hi,
The first two steps are fine (although I'd add "-bin" to the first step just after the "-uthr 5").
I would use randomise to do the statistical analysis though, and not fslmeants with Bonferroni, as randomise gives you more powerful, less conservative, options for multiple comparison correction.
All the best,
Mark
On 26 Jun 2014, at 08:51, Nabi Zorlu <[log in to unmask]> wrote:
> Dear FSL list,
> I planed to have ROI analysis on the tbss output.
>
> I have calculated the ROI with this procedure: for example for splenium of cc.
>
> 1. fslmaths /usr/local/fsl/data/atlases/JHU/JHU-ICBM-labels-1mm.nii.gz -thr 5 -uthr 5 spleniumcc
>
> 2. fslmaths spleniumcc.nii.gz -mas mean_FA_skeleton_mask -bin spleniumccmask
>
> 3. fslmeants -i all_FA_skeletonised -m spleniumccmask.nii.gz
>
> and use ANOVA with Bonferroni correction to compare 3 groups.
>
> is this the appropriate way or should i use randomize to compare 3 groups.
>
> Sincerely,
> Nabi
|