We have run som experiments including DTI on Philips MR-scanners.
Conversion from dicom to nifti have been performed by dcm2nii. In some
cases the conversion have made a .bvec-file containing the orientation
vector, but in some cases there is no .bvec-file. When checking those
series that lack the .bvec-file, some have no
"DiffusionGradientOrientation"-field in the dicom header, other have the
field, but it is always 0. Does anyone have a comment about this? Should
we use another program to get the orientation-vector?