ERROR: talairach_afd: Talairach Transform: transforms/talairach.xfm ***FAILED*** (p=0.0455, pval=0.0034 < threshold=0.0050)

[chrisgorgo]

A user reported issues with talairach transform in recon-all. The issue was resolved by performing rough coregistration of the input images before feeding them to freesurfer:

flirt -dof 6 -in OAS1_bids/sub-0180/anat/sub-0180_T1w.nii.gz -ref average_305.nii.gz -omat rigid.xfm -out OAS1_bids_sub-0180_fixed/sub-0180/anat/sub-0180_T1w.nii.gz

I wonder if we should build this into FMRIPREP - ideally the coregistration would be only saved in the headers so no additional interpolation would be performed.

[hstojic]

I think I get the same error for one of my subjects, using fmriprep 1.2.2:

#@# Talairach Failure Detection Fri Nov 16 01:48:32 UTC 2018
/scratch/scratch/ucjttoj/fnclearning_fmri/dProcessed/freesurfer/sub-s055/mri

 talairach_afd -T 0.005 -xfm transforms/talairach.xfm 

ERROR: talairach_afd: Talairach Transform: transforms/talairach.xfm ***FAILED*** (p=0.0012, pval=0.0000 < threshold=0.0050)

Manual Talairach alignment may be necessary, or
include the -notal-check flag to skip this test,
making sure the -notal-check flag follows -all
or -autorecon1 in the command string.
See:

http://surfer.nmr.mgh.harvard.edu/fswiki/FsTutorial/Talairach

ERROR: Talairach failed!

Linux node-h00a-005.myriad.ucl.ac.uk 3.10.0-693.11.6.el7.x86_64 #1 SMP Thu Dec 28 14:23:39 EST 2017 x86_64 x86_64 x86_64 GNU/Linux

recon-all -s sub-s055 exited with ERRORS at Fri Nov 16 01:48:32 UTC 2018

here is the full crash log: crash-20181116-014833-ucjttoj-autorecon1-2f33a97d-4647-447f-9b29-41dfe5cfc947.txt

Should I do the manual alignment as suggested in the error and then try to run the fmriprep again? Or can I include additional flags in the fmriprep command to resolve it?

[effigies]

I would look at your T1w images (raw, and there should be a preprocessed one in your output directory at this point).

1) If your raw image looks bad, this is a data problem. 2) If your raw image looks good, but your preprocessed looks bad, this is an fMRIPrep problem. 3) If both images look good, this is a FreeSurfer problem. You should finish this step manually, possibly with the assistance of the FreeSurfer mailing list. You can then resume fMRIPrep.

[hstojic]

Thanks a lot for the advice! I just checked both images, and they do look good, so I guess I'll have do it manually.

Is there some stochasticity with that function and worth trying the fmriprep again, or will it for sure be stuck again on the same step?

[effigies]

You can give it a shot again. I don't know whether it's stochastic or not, but trying twice rarely hurts.

[Lindadalaile]

Excuse me,I think I also get the same error for one of my subjects Is there anyone me?I'm crazing Here is the full detail. error.txt

[oesteban]

It doesn't look like an error. At least it doesn't look like this one:

$ grep "Talairach failed" error.txt
$

Please provide further details for us to help.

[Lindadalaile]

It doesn't look like an error. At least it doesn't look like this one:

$ grep "Talairach failed" error.txt
$

Please provide further details for us to help.

Thank you! I had solved this problem with others help. Thank you again.

[effigies]

Oh wait. This wasn't the originator...

[oesteban]

The original thread at neurostars appears as solved now. Even though we didn't directly address this, it seems the prevalence of this problem is low enough to close the issue. I also wonder how much FS7 has improved in this regard.

If anyone experiences this problem again, please reopen.