Dear Michael,
Thank you for sending this. Based on your code, I found that the desired VOI
can be generated without further interactive input using:
xY.name = V.fname;
xY.def = 'mask';
xY.spec = spm_vol(V.fname);
xY.Ic = 0;
[Y,xY] = spm_regions(xSPM,SPM,hReg,xY);
disp(['Saved VOI with ', num2str(size(xY.XYZmm,2)), ' voxels']);
All the best,
Kay
----- Original Message -----
From: "Michael Erb" <[log in to unmask]>
To: "Kay H. Brodersen" <[log in to unmask]>; "SPM-mailbase"
<[log in to unmask]>
Sent: Friday, August 13, 2010 9:24 PM
Subject: Re: [SPM] Extract all voxels belonging to a given cluster
> Hi Kay,
> after selecting a cluster from the results you can write a mask from this
> cluster into the directory of the SPM.mat with the following commands:
> % get coordinates from cursor
> xyz = spm_XYZreg('NearestXYZ',...
> spm_XYZreg('GetCoords',spm_XYZreg('FindReg')),xSPM.XYZmm);
> % find next voxel
> [xyz,ind] = spm_XYZreg('NearestXYZ',xyz,xSPM.XYZmm);
> % cluster indeces
> A=spm_clusters(xSPM.XYZ);
> cXYZ=xSPM.XYZ(:,A==A(ind));
>
> % load mask.ing file
> V=SPM.VM;
> vol=spm_read_vols(V);
> vol(:)=0;
> % set new mask
> for p=1:size(cXYZ,2)
> vol(cXYZ(1,p), cXYZ(2,p), cXYZ(3,p))=1;
> end
> % write mask int SPM.mat directory
> [pt, mn, ext] = fileparts(V.fname);
> if isempty(pt), pt=SPM.swd; end
> V.fname = fullfile(pt,[sprintf('cluster_%0.1f_%0.1f_%0.1f',xyz) ext]);
> V = spm_write_vol(V, vol);
>
> Unfortunately spm_regions seems to be interactive.
>
> Hope this helps,
> Michael
>
>
> Am 13.08.2010 12:10, schrieb Kay H. Brodersen:
>> Dear all,
>>
>> Given an SPM{T} map from a group analysis, I would like to extract the
>> coordinates of the voxels belonging to a particular cluster. I would then
>> like to use these coordinates to extract VOI-eigenvariates in an
>> individual
>> subject. I was wondering what might be the easiest way of accomplishing
>> this?
>>
>> (a) For example, is there a way to select a cluster in a group contrast
>> and
>> save it as a mask? I could then use this mask when calling spm_regions.m.
>>
>> (b) Or is there a way of obtaining the coordinates of all voxels
>> belonging
>> to a certain cluster from xSPM, and then use those to call spm_regions.m?
>>
>> Many thanks for any hints,
>> Kay
>>
>
> --
> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
> Dr. Michael Erb
> Sektion f. experimentelle Kernspinresonanz des ZNS
> Abteilung Neuroradiologie, Universitaetsklinikum
> Hoppe-Seyler_Str. 3, D-72076 Tuebingen
> Tel.: +49(0)7071/2987753 priv. +49(0)7071/61559
> Fax.: +49(0)7071/294371
> e-mail: <[log in to unmask]>
> www: http://www.medizin.uni-tuebingen.de/nrad/sektion/
> <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
>
