[MINC-users] xfm to mat ??

[Soren Christensen]

Hi Ernest,
 I don't think doing this is straigtforward.
The XFM that is estimated using the MNI tools refers to the coordinate
system in your minc file. (Defined by the direction cosines, the
stepsizes and the origo (startpoints) information. When converting to
an Analyze 7.5 file you loose the cosines and the startpoints. As such
I think you would need to apply another xfm and a translation for your
xfm file to work in the Analyze coord system.
Is there a reason you cannot apply the transform (transform and
resample like target) in MINC and then go back to Analyze?
An alternative could be to change the MINC header to reflrect the
analyze coord system (voxel based) prior to co-registration to force
the xfm to refer to this frame of reference.

Soren

On 1/12/07, Ernest Lo <ernestplo at gmail.com> wrote:
> Hello,
>
> Does anyone have experience or knowledge with converting .xfm (for minc)
> files to .mat (for analyze) files?
>
> >From my understanding, the mat files appear to be a simple text file
> containing a 4x4 matrix.  The first three
> columns represent the rotational, scale and shear elements.  The last column
> represents the translational
> components.  The bottom (fourth) row always consists of 0 0 0 1 and makes
> the matrix square.  Therefore
> placing the first three rows of this matrix after the header information of
> an .xfm file should translate this
> .mat into a .xfm useable for minc images.  Does this sound correct?
>
> Thank you,
>
>
> Ernest Lo
> Montreal Neurological Institute
> Montreal, Canada
> _______________________________________________
> MINC-users at bic.mni.mcgill.ca
> http://www.bic.mni.mcgill.ca/mailman/listinfo/minc-users
>