Dear Kushal,
what makes you say that the masks are empty? For example,
Cerebelum_10_L.nii has 144 non-zero voxels:
>> nnz(spm_read_vols(spm_vol('Cerebelum_10_L.nii')))
ans =
144
You could reslice all the mask images at the resolution of your fMRI
data (with nearest neighbour interpolation) with spm_reslice() or you
can leave it to spm_summarise() to do it for you:
>> spm_summarise('fmridata.nii','Cerebelum_10_L.nii')
Best regards,
Guillaume.
On 11/01/16 13:26, kushal veer wrote:
> Hi,
>
> I wants to know that, how we create mask of ROIs of AAL atlas regions ?
>
> I am using the following code to create the mask of regions using AAL
> atlas :
>
> AAL = spm_atlas('load','AAL');
> labels = spm_atlas('select',AAL,'.*');
> for i=1:numel(labels)
> V = spm_atlas('mask',AAL,labels{i});
> V.fname = [labels{i} '.nii'];
> spm_write_vol(V,V.dat);
> end
>
> I obtained the masks image for all the AAL regions.But the all entries in mask images are zero.
>
> Second problem is the dimensions of the mask images do not agree with the dimensions of my original fmri image.
>
> My original fmri images is of dimensions 64 x 64 x 33 x 150 (of single subject)
>
> but the AAL region mask for Cerebelum_10_L (for example) is 91 x 109 x 91.
>
> Shouldn't the mask be the same size as the original image with the appropriate region labeled 1 and the rest labeled 0?
>
> In this case how I multiply these two different dimension image matrix for getting the time series of corresponding AAL atlas region (example Cerebelum_10_L).
>
>
>
> --
> Kushal Veer Singh
> School of Computational & Integrative Sciences
> Jawaharlal Nehru University
> New Delhi - 110067
>
--
Guillaume Flandin, PhD
Wellcome Trust Centre for Neuroimaging
University College London
12 Queen Square
London WC1N 3BG
|