 |
 |
Dear Steve, > I'm trying to understand how to use an explicit mask in SPM8. > > The brainmask.nii in the SPM8/apriori folder has a voxel size of 2mmx2mmx2mm. > My data has a voxel size of 3mmx3mmx3mm. > > 1. Is it a necessity to modify the voxels size of this brainmask.nii image to be the same as my data (i.e. 3mmx3mmx3mm)? Are there advantages to modify the voxels size of the mask? No, you shouldn't need to change the voxel size—SPM should be able to deal with this fine. I can't think of any advantage to reslicing to match the voxel size, though you may get slightly different results around the edge depending on the interpolation method. Alternatively you could make your own brain mask based on segmented structural images (there are some previous posts on this). > 2. In the spm_defaults.m file, there's a line named "defaults.mask.thresh = 0.8;". I don't understand the difference between an explicit mask and that this setting "defaults.mask.thresh". Does the "defaults.mask.thresh" take the brainmask.nii image and transform it into a binary mask based on the transformation i1>0.8 (like you can do with ImCalc) or is it something else? defaults.mask.thresh refers to the masking SPM does during 1st level (timeseries) model estimation, the product of which you can see by viewing the mask.img in a stats directory. If you search the archives for defaults.mask.thresh you should fine some additional information (e.g. https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=SPM;17b284a9.1001). Any explicit masking you specify is applied in addition to this calculated mask. Hope this helps! Best regards, Jonathan -- Dr. Jonathan Peelle Department of Neurology University of Pennsylvania 3 West Gates 3400 Spruce Street Philadelphia, PA 19104 USA http://jonathanpeelle.net/