Submitted by pfuentes on Mon, 06/25/2012 - 15:54
Dear experts,
I am using AlphaSim to determine the minimum cluster size to regard my results as corrected at the cluster level. I would like to know the corrected cluster extent threshold when using quite a low voxel-level threshold, say 0.02 or 0.03 (uncorrected). However, when trying to estimate this extent threshold, AlphaSim crashes after a few iterations giving an error (please, see text below).
It works fine with an uncorrected threshold of 0.01, 0.005 or 0.001. Does anyone have any idea of what might be going wrong? I have a voxel-size of 1mm, 12mm FWHM, and rmm=2.5. Any advice would be greatly appreciated.
I have attached the mask for the stimation of the cluster extent threshold. If anyone could give me the threshold using my mask at a low statistical threshold of p=0.02 and p=0.03, I will appreciate it.
ERROR:
??? Index exceeds matrix dimensions.
Error in ==> rest_AlphaSim at 80
foneimt(numel(find(theCluster==i)))=foneimt(numel(find(theCluster==i)))+1;
Error in ==> rest_AlphaSim_gui>pushbuttonrun_Callback at 55
rest_AlphaSim(mask,outdir,outname,rmm,fwhm,pthr,iter);
Error in ==> gui_mainfcn at 95
feval(varargin{:});
Error in ==> rest_AlphaSim_gui at 27
gui_mainfcn(gui_State, varargin{:});
??? Error using ==> rest_AlphaSim_gui('pushbuttonrun_Callback',gcbo,[],guidata(gcbo))
Index exceeds matrix dimensions.
??? Error while evaluating uicontrol Callback
Best,
Paola.
| Attachment | Size |
|---|---|
 mask_SR.img | 4.41 MB |
| mask_SR.hdr | 348 bytes |
Submitted by YAN Chao-Gan on Fri, 06/29/2012 - 01:55 Permalink
Re
Hi, my guess is the cluster will be too big if you give a too low threshold for individual p level, especially you have voxel-size of 1*1*1.