[MINC-users] dcm2mnc produces xspace:spacing = "irregular"

[Andrew Janke]

>  I'm new to MINC (and to image processing in general) and have encountered
>  the problem herein:
>  http://www2.bic.mni.mcgill.ca/pipermail/minc-users/2006-September/001341.html.
>  The proposed solution appears to work all right but I'm curious as to the
>  nature of the irregularity. What causes dcm2mnc to set the xspace:spacing
>  variable to "irregular"?  Could there be a problem with the original DICOM
>  series? How would I verify that?

In order:   Differing interpretations of the dicom "standard", yes
(and no), easy.

Unfortunately DICOM does not always unambiguously describe the
position of the slices in an image volume (some scanners use
origin/angle, some use absolute position of centre of slices, some use
offsets, etc). What this means is that at times you hit a DICOM
acquisition where the slices might be spaced as such (eg: centres in
Z)

   1.0mm,   2.0mm, 3.0mm, 4.000001mm, 5.0mm, etc

Now we can all guess that the 4.000001 is likely a rounding error and
that this acquisition is indeed 1mm spaced, but in code we have to
define this "epsilon" value. In dcm2mnc's case it is set pretty low so
a few will slip through the cracks as "irregular" when they in all
reality likely were not.

As for how to test this just run dcm2mnc -debug and you will see all
the slice co-ordinate information.


a