Hi list,

trying to import fmri dicom files from HDx GE scanner but no success as slice order gets messed up. Mess is almost like every other slice would have been swapped but that is not completely the pattern that I see after to3d.

to3d -time:zt 28 250 1764 alt+z '/data/MR.1.2.840.113619.2.207.3596.11798700.24698.118277622*'
++ to3d: AFNI version=AFNI_2007_05_29_1644 (Sep 18 2007) [64-bit]
++ Authored by: RW Cox
*+ WARNING: TR expressed in milliseconds will be converted to TR=1.764000s
++ Counting images: total=7000 2D slices
++ DICOM WARNING: VR mismatch in element (0028,0106)
++ Data detected to be oblique
++ Each 2D slice is 64 X 64 pixels
++ Voxel dimensions: 4.0000 X 4.0000 X 4.4000 mm
++ Image data type = short
++ Reading images: ..........................................
*+ WARNING: Image Positions do not lie in same direction as cross product vector - 1.000000
++ Command line TR=1764ms ; Images TR=1.764s
++ Making widgets.....

Then there is another issue with 3dretroicor and slice information in NIFTI files (converted using mri_convert of Freesurfer): I define proper slice timing information using nifti_tool but 3dretroicor obviously does not recognize slice info from .nii-files as it does not matter what I define, written nii-file from 3dretroicor does not have any info regarding slices (slice_code etc.). I tested 3dretroicor for data with different slice settings (alt+z & alt-z) in to3d (see above) and it did matter for 3dretroicor output.

So, to3d fails for our data and 3dretroicor does not like our NIFTI-files, helpful comments highly appreciated! Hope problem description was detailed enough.

BR, Tuomo