I have diffusion data that is reverse phase encoded for all directions. When I run this with the -rpe_all option it works, but the topup step takes quite a long time. To reduce the time cost of this step, I am trying to use -se_epi to specify an image that contains 2 b0s (one in each direction). This returns: dwifslpreproc: [ERROR] If explicitly including SE EPI images when using -rpe_all option, they must come with their own associated phase-encoding information in the image header. How do I ensure that my se epi specified imaged come with their own associated phase-encoding information in the image header?
an update: I was able to specify the phase encoding direction for each of my b0 pair using:
mrconvert out_dir/b0_AP.nii.gz out_dir/b0AP.mif -set_property PhaseEncodingDirection j
and -j for the PA b0
now there is an error reading the file topup_datain.txt
The log reads:
mrconvert se_epi.mif -coord 2 56 - | mrcat se_epi.mif - se_epi_pad2.mif -axis 2
mrconvert dwi.mif -coord 2 56 - | mrcat dwi.mif - dwi_pad2.mif -axis 2
mrconvert se_epi_pad2.mif topup_in.nii -strides -1,+2,+3,+4 -export_pe_table topup_datain.txt
and when I check the file topup_datain.txt:
0 1 0
0 -1 0
this appears to be missing the readout time and the second row has a space before the first zero. Any ideas why this error may be occurring? Or how to fix? Thanks in advance for any help.
JG