Hi Mark
No need to look at my last question: I found out what I did wrong, it used stage 3 results instead of stage 2!
Thanks for all the help
Best
Heidi
================================================
Dr. Heidi Jacobs
Postdoc researcher
Faculty of Health, Medicine and Life Sciences
School for Mental Health and Neurosciences
Division Cognitive Neuropsychiatry and Clinical Neurosciences
Alzheimer Center Limburg
[log in to unmask]
www.maastrichtuniversity.nl
www.heidijacobs.nl
Dr. Tanslaan 12, 6229 ET Maastricht
P.O. Box 616, 6200 MD Maastricht, The Netherlands
T +31 43 38 84 090 F +31 43 38 84 092
================================================
________________________________________
Van: FSL - FMRIB's Software Library [[log in to unmask]] namens Mark Jenkinson [[log in to unmask]]
Verzonden: dinsdag 8 januari 2013 11:18
Aan: [log in to unmask]
Onderwerp: Re: [FSL] using cluster, fslmaths, fslstats or fslmeants???
Dear Heidi,
You can use the tool "cluster" to create an image where each individual cluster gets a unique number, and then split these up into separate masks with fslmaths.
For the cluster command, you do what you wrote except that you use "--thresh=0.95" (it is essential to use the equals sign with double-minus options, and have no spaces), you don't need the --connectivity option, but you do need an output image, via the --oindex=someimagename. This output image is the one that has a unique number associated with each cluster.
Once you've done this you can extract the individual clusters by using fslmaths. For example, say you want to extract the cluster that has the value 3 (i.e. all voxels within this cluster are labelled with the value 3 in the image "someimagename" above). So you can do this with:
fslmaths someimagename -thr 3 -uthr 3 -bin cluster3mask
This output mask (cluster3mask) can then be used with fslmeants or fslstats or atlasquery to extract properties associated with just this cluster.
I hope this helps.
All the best,
Mark
On 7 Jan 2013, at 15:36, Jacobs H (NP) <[log in to unmask]> wrote:
> Hi Folks,
>
> Looking at the results of my resting state analyses, I saw group differences in several components, which make me happy of course.
> Now I would like to create clusters from these results.
> For example: I see a few voxels in the hippocampus and several voxels in the precuneus. How can I make two clusters from this (one for the hippocampus and one for the precuneus)?
> And how can I extract the values from these clusters for graphical purposes?
>
> Encouraged by Steve Smith,I looked into my course materials again, but I am still confused on the exact order of using the commands.
> Would this be the correct procedure:
>
> 1. First make a file with only the clusters: cluster -i <DR3_stage_corpstatfile> --thresh 0.95 (should I also enter --connectvitiy??)
> 2. Extract the values of this: fslmeants -i <DR3_stage_corpstatfile> -m <cluster_outputfile> -0 <new output>
>
> But I am confused on how to separate the different clusters in the DR3stage file? Should I do that with fslmaths or fslstats?
> Any help would be appreciated!
>
> Thank you
> Heidi
>
> ================================================
> Dr. Heidi Jacobs
> Postdoc researcher
> Faculty of Health, Medicine and Life Sciences
> School for Mental Health and Neurosciences
> Division Cognitive Neuropsychiatry and Clinical Neurosciences
> Alzheimer Center Limburg
> [log in to unmask]
> www.maastrichtuniversity.nl
> www.heidijacobs.nl
>
> Dr. Tanslaan 12, 6229 ET Maastricht
> P.O. Box 616, 6200 MD Maastricht, The Netherlands
> T +31 43 38 84 090 F +31 43 38 84 092
> ================================================
|