GM and CSF probsegs misnamed

[lundq163]

What happened?

When running some data through the --anat-only version of the pipeline, I found that the GM and CSF probsegs were oftentimes misnamed, with the GM being named as CSF and vice versa. This issue was irrespective of the space the derivatives were in as well.

What command did you use?

singularity run --cleanenv \
-B ${fmriprep_dir}/bids_dir/sub-01_ses-01:/bids_dir \
-B ${fmriprep_dir}/processed/fmriprep/sub-01_ses-01:/output_dir \
-B ${fmriprep_dir}/work_dir/fmriprep/sub-01_ses-01:/wd \
-B ${run_dir}/license.txt:/license.txt \
/home/faird/shared/code/external/pipelines/fmriprep/fmriprep_24.1.0.sif \
--output-spaces MNI152NLin6Asym:res-2 \
--fs-license-file /license.txt \
--project-goodvoxels \
--omp-nthreads 3 \
--cifti-output 91k \
--anat-only \
--skip_bids_validation \
-vv \
-w /wd \
/bids_dir /output_dir participant

What version of fMRIPrep are you running?

24.1.0

How are you running fMRIPrep?

Singularity

Is your data BIDS valid?

Yes

Are you reusing any previously computed results?

No

Please copy and paste any relevant log output.

No response

Additional information / screenshots

I think my data is BIDS valid. I am just using two separate MPRAGE nifits, but I relabeled them to be T1w run-01 and run-02. I used --skip_bids_validation though still to quickly/efficiently produce the derivatives I needed. I am wondering if this style of data may have had an effect on the masking of the cortical areas.

[effigies]

I think this is a duplicate of https://github.com/nipreps/fmriprep/issues/3195. Would you mind having a look to confirm?

[lundq163]

yes, can confirm the issues look quite similar. My assumption is that the MPRAGE niftis I am calling T1s have "different tissue classes are too uniform for FAST to be able to accurately model them", per @pauldmccarthy 's comment on that other issue. Feel free to close this issue and I am happy to monitor #3195 and run any tests if needed.