HCP-EP Diffusion Processing

[Matthew Danyluik]

Hi there - 

I'm using the HCP Pipelines, v5.0.0, to process diffusion data from the HCP Early Psychosis dataset. Processing fails, and in the error log file, I get the following warning printed several times:

WARNING:: Inconsistent orientations for individual images when attempting to merge.
          Merge will use voxel-based orientation which is probably incorrect - *PLEASE CHECK*!

And then a series of messages indicating that run_eddy.sh failed, with no further information as to why.

My best guess is that this is due to slight differences in the FOV between the dir99 and dir107 acquisitions as indicated in the scanning protocol, though I don't know enough to say if that's the problem causing this error, or how it might be fixed.

Any suggestions are welcome. I'm happy to attach any log files or exact commands run if you'd like. Though I will note that, outside of setting dataset-specific parameters in DiffusionPreprocessingBatch.sh, I'm using strictly default settings (besides using the CPU version of eddy).

Thank you in advance for your help.

[Harms, Michael]

That particular WARNING is unlikely to be related to the real cause of the failure.  Did you check the log for the word "error"?  Or attach the log.

 

-- 

Michael Harms, Ph.D.

-----------------------------------------------------------

Professor of Psychiatry

Washington University School of Medicine

Department of Psychiatry, Box 8134

660 South Euclid Ave.                        Tel: 314-747-6173

St. Louis, MO  63110                          Email: mha...@wustl.edu

--
You received this message because you are subscribed to the Google Groups "HCP-Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to hcp-users+...@humanconnectome.org.
To view this discussion visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/263d9cfb-62a5-4ef1-9b88-8f4c0a790a5an%40humanconnectome.org.

 

---

The materials in this message are private and may contain Protected Healthcare Information or other information of a sensitive nature. If you are not the intended recipient, be advised that any unauthorized use, disclosure, copying or the taking of any action in reliance on the contents of this information is strictly prohibited. If you have received this email in error, please immediately notify the sender via telephone or return mail.

[Matthew Danyluik]

Hi Michael - 

Thanks very much for the reply. I'm attaching both the .o and .e log files here. 

[Harms, Michael]

Hmm.  You need to check the .o file in a situation like this.

 

That said, the report of the actual cause of the error appears corrupted for some reason:

 

Fri Oct 17 15:06:28 EDT 2025:run_eddy.sh: /cvmfs/restricted.computecanada.ca/easybuild/software/2023/x86-64-v3/Core/fsl/6.0.7.7/bin/eddy_cpu --cnr_maps --imain=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/Pos_Neg --mask=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/nodif_brain_mask --index=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/index.txt --acqp=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/acqparams.txt --bvecs=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/Pos_Neg.bvecs --bvals=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/Pos_Neg.bvals --fwhm=0 --topup=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/topup/topup_Pos_Neg_b0 --out=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/eddy_unwarped_images

=Q�<,V

0��b

EDDY::: Eddy failed with message X�<,V

Fri Oct 17 15:06:41 EDT 2025:run_eddy.sh: While running '/scratch/danyluik/hcp-ep/scripts/HCPpipelines-4.8.0/DiffusionPreprocessing/scripts/run_eddy.sh --nvoxhp= --ff= --wss=0 --repol=0 --sep-offs-move=0 --rms=0 --ol_nstd= --gpu=0 --cuda-version= --workingdir=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy --peas=1 --fwhm=0 --resamp=':

Fri Oct 17 15:06:41 EDT 2025:run_eddy.sh: ERROR: '/cvmfs/restricted.computecanada.ca/easybuild/software/2023/x86-64-v3/Core/fsl/6.0.7.7/bin/eddy_cpu --cnr_maps --imain=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/Pos_Neg --mask=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/nodif_brain_mask --index=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/index.txt --acqp=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/acqparams.txt --bvecs=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/Pos_Neg.bvecs --bvals=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/Pos_Neg.bvals --fwhm=0 --topup=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/topup/topup_Pos_Neg_b0 --out=/scratch/danyluik/hcp-ep/imagingcollection01/1006_01_MR/Diffusion/eddy/eddy_unwarped_images' command failed with return code: 1

 

 

Could you try a different subject?  Or try to use a GPU instead?  Either of those might provide a clue.

 

Cheers,

-MH

[Matthew Danyluik]

Hi Michael - 

I tried a different subject and the .o file was corrupted there as well. I'll also point out that I was able to run diffusion processing successfully using the same setup for the HCP-Development dataset, as well as the 5 subjects in HCP-Early Psychosis with dir98 and dir99 acquisitions instead of dir99 and dir107 (this didn't seem to be a trivial effect of file name).

If you think running on a GPU might give a clue, I can try that, though I'd rather keep my workflow to CPUs if at all possible (I ran into issues with resource usage on our cluster once I scaled to an entire dataset with the GPU-enabled version of eddy).

Cheers, and thanks so much for the help with this.

[Tim Coalson]

I don't think there was any file corruption, I remember a report where eddy output a nonsense message like that.  Unfortunately, I don't know that the underlying cause was found (or a solution):

https://github.com/Washington-University/HCPpipelines/issues/340

Tim

To view this discussion visit https://groups.google.com/a/humanconnectome.org/d/msgid/hcp-users/804ba750-80c9-447b-ae35-e2ae6e7fc04fn%40humanconnectome.org.

[Matthew Danyluik]

Hi Tim and Mike - 

Maybe just to wrap up the discussion, I wanted to let you guys know that, after upgrading to FSL 6.0.7.18 from 6.0.7.7, the log files became much more interpretable.

As it turns out, eddy was interpreting the data as "non-shelled", which can be corrected by passing a --data-is-shelled flag. I'm still unclear on why this happened, but in any event, it's more a question for FSL support or the HCP-EP team.

Thanks for your support,

Matt.