[MINC-users] error running mincmath

[Andrew Janke]

On 03/06/05, Simon Fristed Eskildsen <se@hst.aau.dk> wrote:
> Package MNI N3
> 
> Command: nu_correct native/t1_native.mnc.gz temp/t1_nuc.mnc -clobber
> -keeptmp
> 
> Error message:
> Files /usr/tmp/nu_correct_28314//t1_nuc_field.mnc and
> native/t1_native.mnc.gz have different coordinate spacings (zspace)
> nu_evaluate: crashed while running mincmath (termination status=256)
> nu_correct: crashed while running nu_evaluate (termination status=256)
> 
> Same output on mincinfo:
> image: signed__ short -32768 to 32767
> image dimensions: zspace yspace xspace
>     dimension name         length         step        start
>     --------------         ------         ----        -----
>     zspace                    160            1     -30.4761
>     yspace                    256      -0.9375      148.799
>     xspace                    256      -0.9375      119.531
> 
> Same direction cosines:
> mincheader /usr/tmp/nu_correct_28314//t1_nuc_field.mnc | grep
> direction_cosine
>                 zspace:direction_cosines = 0., 0., 1. ;
>                 yspace:direction_cosines = -0., 1., -0. ;
>                 xspace:direction_cosines = 1., -0., -0. ;
> 
> Maybe the -0 entries in the cosines are causing the problem?
> This is the first time I have encountered this problem. The input minc
> file have been converted from dicom using dcm2mnc (minc-1.4, netcdf -3.5.1)
> 
> Any ideas?

Another thing to remember is that mincinfo only reports starts and
steps with a limited number of decimal places (and thus can round
results).  To check if the two files are indeed identical with repect
to step sizes use something like this:

   mincinfo -attvalue zspace:start in.mnc
   mincinfo -attvalue zspace:start two.mnc

The simple fix to this is as louis suggested, simply run mincmath with
the -nocheck_dimensions argument.

-- 
Andrew Janke      (a.janke@gmail.com || www.cmr.uq.edu.au/~rotor)
Australia->Brisbane     H: +61 7 3390 6332  || M: +61 4 2138 8581