Fmriprep discards t1 and produces no output of anat

1

Hello all,

Me and my colleague are performing preprocessing on multiple subjects and multiple sessions per subject, with anatomical and functional scans.
Unfortunately, on certain subjects, fmriprep does not produce anat output. In the html file, it reports discarding the t1 scans while I do not see why. The t1s seem correct when inspecting in fsleyes.

I used the following script:
singularity run --cleanenv -B /dbstore/Rijnstate/SYNAPSE/BIDS2:/BIDS2 /dbstore/Rijnstate/fmriprep-20.2.6.simg
/BIDS2 /BIDS2/derivatives participant --participant_label $sub
–skip_bids_validation
–longitudinal
–fs-license-file /BIDS2/license.txt --fs-no-reconall --use-aroma

I do not know if it is a relevant issue, but I have also noticed that some jsons of func scans report “RepetitionTimeExcitation” next to other jsons which only report the “RepetitionTime”. In the jsons which also report the former, the RepetitionTime reports a very long TR of around 3sec, while it should be 1.9 (which is then reported in “RepetitionTimeExcitation”). The correct TR is then mentioned in “RepetitionTimeExcitation”. Again, I doubt this is a related issue, but nonetheless maybe relevant to mention.

Does someone know what could be the issue of FMRIprep discarding t1s?

Thanks in advance,
Joey

2

A few things that could help us help you:

  1. Try updating to 21.0.0 and see if the error persists
  2. Is the dataset BIDS-valid?
  3. In you BIDS directory, can you paste the output of tree sub-EXAMPLESUB so we can see what the file structure for one subject looks like (fill in EXAMPLESUB with a subject name).
  4. I think --longitudinal mainly changes how FreeSurfer creates a template, so I am not sure if it is compatible with --fs-no-reconall. Can you try including FreeSurfer in your runs instead, unless do you have another reason to not use it?

Hope we can get to the bottom of this,
Steven

3

Thank you for your help Steven. I have been trying to figure out why, but the problem has luckily been solved. I have upgraded to v21 and since have not encountered the problem above.