[ITK-users] dicom viewer/nifti discrepancy

[Nicholas Tustison]

Hi all,

I have a set of dicom data that were annotated by clinical
folk.  Specifically, I have slice numbers for each image 
volume which indicate a particular feature of interest.  The 
problem is that the slice number can either be counted from
the most inferior slice or the most superior slice.  When I
view these dicom image volumes in Osirix or Fiji, the slice
number provided by the viewer corresponds to my clinical 
numbers.  However, when I convert to nifti (using dcm2nii or 
mriconvert), the index, in ITK terminology, always starts from the 
most inferior slice.  At first, I thought it was some dicom tag that 
was being read by the viewer to orient the index but, after following 
a couple of possibilities, I haven’t found anything.  Has anyone
come across anything like this?

Thanks,
Nick