issue saving result "Coreg T1 to Func"

Submitted by orsolini on

firstly I would like to thank you for these great tools.
my issue:
Using DPARSF "Coreg T1 to Func" is computed without errors but then the result is not saved in T1ImgCoreg.
Only the result of "crop T1" is saved correctly.
I tried to do the the same elaboration with SPM8 and the resulting coregistered file is correctly saved.
Looks like it's missing a call to a write-file routine around line 1096 in DPARSF_run. Or I am missing something?
I'm using: DPARSF_V2.2_121225.


YAN Chao-Gan

Wed, 02/20/2013 - 01:09

For "Coreg T1 to Func", it only changes the affine matrix in the header. Thus you will not see new files, but you will see the files are changed if you look at the modification date.



Just because I needed the coreg image file I modified the jobmats/Coregister.mat field named "estimate" in "estwrite" for SPM8 to also write the results.
I'm not considering this a correction, just something useful to me, don't know if this would be of any utility for other.
I'm pulling workarounds for my issues on GitHub, I'd love to give my small contribution to your project.

kind regards

Hi Orsolini,

Thanks a lot for your contribution, your efforts will make this work better!

Regarding to estimate or estwrite, this is not a bug but an intended design. DPARSF coregisters the T1 image to the functional image WITHOUT reslice. However, the orientation of the T1 image is aligned with functional data since a rigid-body coregistration is performed here and the information is stored in the header of T1 image. There is no need to reslice the T1 image in a resolution of functional images (from 1*1*1 to 3*3*3 typically). If you have a special need, you may use estimate and write.

In addition, thanks you revision for the bug in creating mean functional image for 4D files.