Hi
On 16 Mar 2010, at 15:22, Nathan Johnson wrote:
> Hello Experts,
>
> I have sifted through the archives searching for answers to these
> questions with minimal luck. Therefore I must ask a couple of
> questions. In an attempt to identify significant clusters in my DTI
> data (after running Randomise) I have had some difficulty. I have
> created a cluster map and .txt file using the following command:
> cluster --in=tbss_vox_p_tstat4.nii.gz --thresh=0.95 --
> olmax=output.txt --othresh=output_image
You should normally use corrected p-values, not uncorrected.
> After looking through the text file and image I can identify the
> clusters that I am interested in, however I am having a problem
> isolating my cluster of interest so that I can make a mask. I use
> the following command: fslmaths output_image -thr 1064.5 -uthr
> 1065.5 -bin cluster_1065, but when I try to load the image nothing
> comes up and I cannot use this cluster as a mask. I would greatly
> appreciate any suggestions.
Maybe you need to adjust the min and max intensity display values in
FSLView in order to see this. It may be very small.....
> I have used the following steps with some results, however when I
> mask with the entire region (superior_longitudinal_fasciculus_R) I
> get really low FA values. I assume this is because many of the
> voxels within the mask are close to zero?
Quite likely, yes.
Cheers.
>
> 1. fslmaths tbss_vox_p_tstat1 -thr 0.95 -bin mymask
> 2. fslmaths mymask -mas superior_longitudinal_fasciculus_R
> RSLF_mask_95
> 3. fslmeants -i all_FA -m RSLF_mask_95 -o RSLF_FA
>
> If someone can help me get from Randomise to the fslmeants command
> above (step #3) I would greatly appreciate it.
>
> Thank you,
>
> Nate
>
>
>
---------------------------------------------------------------------------
Stephen M. Smith, Professor of Biomedical Engineering
Associate Director, Oxford University FMRIB Centre
FMRIB, JR Hospital, Headington, Oxford OX3 9DU, UK
+44 (0) 1865 222726 (fax 222717)
[log in to unmask] http://www.fmrib.ox.ac.uk/~steve
---------------------------------------------------------------------------
|