I am comparing various options for normalization of lesioned brains across ants, fsl and spm12, and I am confused by the following result:

All of the warped images (w*) that spm12 generates are not interpretable by fslstats (although you can view the image in fsleyes and see the brain and range of values).  What I mean by "not interpretable" is that fslstats is reporting all the values in the image are nan. fslmaths also cannot operate on these images as it thinks they are all nans.

FSL does understand the values in the deformation field and the segmentation images, just not the warped images...I have bolded the results below for all the spm images...

I have linked to a brain image: https://arizona.box.com/s/8kxixwhpo97dmg7et96hw6llgyfbkbic so you can see the problem: it can be viewed with fsleyes but fslstats and fslmaths think it contains only nans.

I am working with Matlab 2016b, spm12 and Mac OSX Sierra.  Any insights would be greatly appreciated!

Thanks,

Dianne