Print

Print


Hi Matthew,

thank you very much for the response.

The lmax_zstat3.txt (and the same for my other contrasts) provides the
cluster index, z- value, and x/y/z coordinates. However, the voxel size of
each of these local maxima clusters still seems to be missing (for example,
I have 5 local maximas falling under the cluster index of 5, and I'd like
the cluster size of each). When I look up the main cluster table
(cluster_zstat3_std.txt; the same from the html provided by FEAT), the
voxel size is only listed for each cluster index:

Cluster Index        Voxels         ....
5                            5020
4                           1240
3                            1101
2                            989
1                            345

but each cluster index, has multiple local maxima listed on
the cluster_zstat3_std.html, and it is with getting the voxel size of each
of these local maximas that I am having trouble with. Thank you again.

On Wed, Jan 6, 2016 at 7:56 AM, Matthew Webster <
[log in to unmask]> wrote:

> Hello Paul,
>                     The local maxima information ( e.g. in lmax_zstat3.txt
> ) will give a cluster index for each maxima - this can then be looked up in
> the main cluster table output.
>
> Hope this helps,
> Kind regards
> Matthew
>
>
> Greetings FSL community,
>
> sorry for reposting a question, but I'm having issues being able to report
> the voxel size for each cluster index of the local maxima reported from a
> FEAT analysis. I've done the FEAT analysis, and would like to report the
> local maxima, their associated regions of activity based on their
> coordinates, and voxel size. I've been using the cluster command with the
> option to output the local maxima (number = 5) text file:
>
> cluster -i thresh_zstat3 -t 2.3  -n 5 --mm --olmax=lmax_zstat3.txt
>
> the output matches that from the FEAT report, but not the local maxima
> information. I was hoping to report the cluster size (voxels) of each local
> maxima found in each cluster index (if that makes sense...). I feel like
> I'm missing something easy, but not been able to get the output I was
> hoping for.
>
> Any help with this would be greatly appreciated. Thank you.
>
>
>