Strange distortion correction

Software Support
,
1

Summary of what happened:

I am going through the outputs of fmriprep and am noticing that the distortion correction (based on B0 fieldmaps) looks strange. Any insights on what might be happening would be appreciated.

Command used (and if a helper script was used, a link to the helper script or the command generated):

singularity run --cleanenv \\
-B /home/hfluhr/shares-hare/ds-learning-habits:/data \\
-B /scratch/hfluhr/workflows/ds-learninghabits:/work \\
-B /home/hfluhr/shares-hare/ds-learning-habits/derivatives:/out \\
/home/hfluhr/data/containers/fmriprep24 /data /out/fmriprep-24.0.1_new participant --participant-label sub-09 -w /work

Version:

24.0.1

Environment (Docker, Singularity / Apptainer, custom installation):

I am running with Singularity.

Data formatted according to a validatable standard? Please provide the output of the validator:

bids-validator@1.14.6
[33m1: [WARN] The recommended file /README is very small. Please consider expanding it with additional information about the dataset. (code: 213 - README_FILE_SMALL)e[39m
		./README
[36m	Please visit https://neurostars.org/search?q=README_FILE_SMALL for existing conversations about this issue.e[39m
[34me[4mSummary:e[24me[39m                  e[34me[4mAvailable Tasks:e[24me[39m        e[34me[4mAvailable Modalities:e[39me[24m 
        2203 Files, 50.9GB        Reward Pairs            MRI                   
        71 - Subjects                                                           
        1 - Session                                                             

[36m	If you have any questions, please post on https://neurostars.org/tags/bids.e[39m

Relevant log outputs (up to 20 lines):

240815-14:02:50,644 nipype.workflow IMPORTANT:
	 fMRIPrep finished successfully!

Screenshots / relevant information:

From the html report for a sample subject:

Summary
* Original orientation: LPS
* Repetition time (TR): 2.33s
* Phase-encoding (PE) direction: Right-Left
* Single-echo EPI sequence.
* Slice timing correction: Applied
* Susceptibility distortion correction: FMB (fieldmap-based) - phase-difference map
* Registration: FreeSurfer `bbregister` (boundary-based registration, BBR) - 6 dof
* Non-steady-state volumes: 0

Corrected:

image
image1920×1016 126 KB

Distorted:

image
image1920×1017 133 KB

2

Hi @hugofluhr,

Can you try --ignore fmap-jacobian?

Best,
Steven

3

Hi @Steven,

Here are the same figures after running with --ignore fmap-jacobian. I don’t see much of a difference
Distorted:

image
image1882×1034 130 KB

Corrected:
image
image1878×1034 128 KB

Thanks,
Hugo

4

@hugofluhr I get a similar pattern of results on a diffusion dataset that is oriented PSR. When I reorient the original data to LAS the susceptibility distortion results come out much better, though I have not been able to figure out why.

5

@jhau thanks for contributing to this thread!

I realized that I had messed up my bids conversion so the above results are not valid (the wrong B0 was used for the bold run).
I will open another thread as I am still facing SDC issues after having fixed the bids conversion.