Hello again,
now I've tried to obtain cluster statistics for a T image from a GLM flex model with the following (I want to use an initial uncorrected voxel threshold of .001 and a cluster threshold of .05 FWE or .05 FDR). I specified FWHM based on the values stored in oo.mat, but not RESELS (didn't find anything related).
mapparameters.type='T';
mapparameters.df1=27;
mapparameters.thresh=.001;
mapparameters.label='aal_MNI_V4';
mapparameters.voxlimit=1000;
mapparameters.SPM=0;
mapparameters.separation=8;
mapparameters.nearest=0;
mapparameters.cluster=50;
mapparameters.conn=18;
mapparameters.mask=[];
mapparameters.sign='pos';
mapparameters.out=[];
mapparameters.UID='test';
mapparameters.threshc=.05;
mapparameters.FWHM=[6.12377069382456 6.13072537962399 5.88209396756923];
mapparameters.savecorrected.type='cFWE';
[voxels voxstats cluststats sigthresh regions mapparameters UID] =peak_nii('0001_T_Effect.nii',mapparameters);
which results in
Using the default of no RESELS. FWHM was incorrectly specified or missing.
Threshold is:3.421
I also don't get any mat files then, but variables in workspace. clusterstats is empty. In sigthresh I have values for FDRp, FDRpk18, FDRpk26, FWEc and FDRc though. As far as I understood it should be possible to obtain cluster p values for each of the clusters? So what have I done wrong?
Best,
Helmut
