You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
My understanding is that the tensor dimensions were messed up when the warp field data was loaded from the HDF5 file, somehow the first and the third dimensions were swapped.
ants_t1_to_mniComposite.h5 was from a working directory of fMRIPrep 21.0.2, and it's identical to sub-*_from-T1w_to-MNI152NLin2009cAsym_mode-image_xfm.h5 in the output directory. The NIFTI file was generated using CompositeTransformUtil --disassemble ants_t1_to_mniComposite.h5 ants_t1_to_mniComposite.