[Insight-users] Regarding Resampling using image orientation patient and image position patient

[Sachin Jambawalikar]

Hi Luis and Jolinda,

I'm trying to resample  one volume series to another volume series based on the 
transformation matrix formed by the direction cosines and the image
position patient  for the  left most point of the first slice in each
volume.

I find the inverse of 4x4 affine matrix(obtained by direction cosines
+ image postion patient) for the fixed volume and  premultiply it to
the transformation of the moving volume. (both volumes are different
series of the same dicom study).

In resample image filter I set the  tranform matrix as the 9
(rotation+shear +scaling) components  of the affine transform matrix
and the transform vector is set as the 3 translation coordinates  .

After resampling  all I get is a volume of zeros  with correct
dimensions and spacing of the fixed volume.

Is there a way in itk to  resample  moving volume to  fixed volume
based on the dicom  header information ??

Regards
--Sachin