Lesion mask potentially not used (fmriprep 23.0.2)

AntoninHUDRY · October 7, 2025, 10:53am

Summary of what happened:

I’m very new to the fmri field, sorry in advance if I’ve done some silly mistakes.

I’m running fmriprep on a single stroke patient with one session that includes FLAIR, T2w, and T1w scans, three functional tasks, and fieldmaps (magnitude1, magnitude2, phasediff). The preprocessing runs without errors, but I suspect that the lesion mask I provide is not being used. The mask is:

in the subject’s anat/ folder
binary and in the same space and orientation as the T1w image
named with the _roi suffix
listed in .bidsignore
validated successfully by the bids validator

However, the following citation line in the citation.html file does not appear: “with cost function masking with the lesion mask to minimize warping of healthy tissue into damaged areas (Brett et al., 2001).”
Likewise, the subject HTML report does not mention the lesion mask, and a downstream pipeline (fmristroke) that depends on fmriprep derivatives from versions <23.2 is reporting missing components.

I’ve tried several variations, running without session folders, adding a JSON sidecar for the lesion mask, forcing BBR and skull stripping, but the outcome hasn’t changed. I’ve included below the fmriprep outputs and a screenshot of my folder organization.

Am I right to think that the lesion mask is not used? If so, what should I check next to ensure that fmriprep correctly recognizes and applies the lesion mask during normalization?

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

runai submit --name fmriprep-test23-0-2 ...<runai commands>... -i .../fmriprep:23.0.2 -- /data/data/RevS /data/data/RevS/derivatives/fmriprep_v23_0_2 participant --fs-license-file /data/data/license_freesurfer.txt --work-dir data/data/RevS_work --verbose --output-layout bids --write-graph --return-all-components

Version:

fmriprep v23.0.2

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

Docker

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

fmriprep validator only warned about authors list and readme being small.

Relevant log outputs (up to 20 lines):

fmristroke output:

251007-08:54:53,438 nipype.workflow WARNING:
A ttempted to access pre-existing bold derivatives at </data/data/RevS/derivatives/fmriprep_v23_0_2_no_ses>, however not all expectations of fMRIPrep were met (for participant <RevS01>.
251007-08:54:53,438 nipype.workflow WARNING:
A ttempted to access pre-existing bold derivatives at </data/data/RevS/derivatives/fmriprep_v23_0_2_no_ses>, however not all expectations of fMRIPrep were met (for participant <RevS01>.

3 times the same error in nipype (used by fmristroke), which I suspect is caused by missing files

Traceback:
Traceback (most recent call last):
File "/usr/local/lib/python3.10/dist-packages/nipype/interfaces/base/core.py", line 397, in run
runtime = self._run_interface(runtime)
File "/usr/local/lib/python3.10/dist-packages/nipype/interfaces/utility/wrappers.py", line 142, in _run_interface
out = function_handle(**args)
TypeError: _group_outputs() missing 1 required positional argument: 'in_outputs'

Screenshots / relevant information:

Folder architecture:

Output of fmriprep:

sub-RevS01_ses-T1_acq-mprage_desc-aparcaseg_dseg.nii.gz                         sub-RevS01_ses-T1_acq-mprage_hemi-L_thickness.shape.gii
sub-RevS01_ses-T1_acq-mprage_desc-aseg_dseg.nii.gz                              sub-RevS01_ses-T1_acq-mprage_hemi-R_curv.shape.gii
sub-RevS01_ses-T1_acq-mprage_desc-brain_mask.json                               sub-RevS01_ses-T1_acq-mprage_hemi-R_inflated.surf.gii
sub-RevS01_ses-T1_acq-mprage_desc-brain_mask.nii.gz                             sub-RevS01_ses-T1_acq-mprage_hemi-R_midthickness.surf.gii
sub-RevS01_ses-T1_acq-mprage_desc-preproc_T1w.json                              sub-RevS01_ses-T1_acq-mprage_hemi-R_pial.surf.gii
sub-RevS01_ses-T1_acq-mprage_desc-preproc_T1w.nii.gz                            sub-RevS01_ses-T1_acq-mprage_hemi-R_smoothwm.surf.gii
sub-RevS01_ses-T1_acq-mprage_desc-preproc_T2w.json                              sub-RevS01_ses-T1_acq-mprage_hemi-R_sulc.shape.gii
sub-RevS01_ses-T1_acq-mprage_desc-preproc_T2w.nii.gz                            sub-RevS01_ses-T1_acq-mprage_hemi-R_thickness.shape.gii
sub-RevS01_ses-T1_acq-mprage_desc-ribbon_mask.nii.gz                            sub-RevS01_ses-T1_acq-mprage_label-CSF_probseg.nii.gz
sub-RevS01_ses-T1_acq-mprage_dseg.nii.gz                                        sub-RevS01_ses-T1_acq-mprage_label-GM_probseg.nii.gz
sub-RevS01_ses-T1_acq-mprage_from-fsnative_to-T1w_mode-image_xfm.txt            sub-RevS01_ses-T1_acq-mprage_label-WM_probseg.nii.gz
sub-RevS01_ses-T1_acq-mprage_from-MNI152NLin2009cAsym_to-T1w_mode-image_xfm.h5  sub-RevS01_ses-T1_acq-mprage_space-MNI152NLin2009cAsym_desc-brain_mask.json
sub-RevS01_ses-T1_acq-mprage_from-T1w_to-fsnative_mode-image_xfm.txt            sub-RevS01_ses-T1_acq-mprage_space-MNI152NLin2009cAsym_desc-brain_mask.nii.gz
sub-RevS01_ses-T1_acq-mprage_from-T1w_to-MNI152NLin2009cAsym_mode-image_xfm.h5  sub-RevS01_ses-T1_acq-mprage_space-MNI152NLin2009cAsym_desc-preproc_T1w.json
sub-RevS01_ses-T1_acq-mprage_hemi-L_curv.shape.gii                              sub-RevS01_ses-T1_acq-mprage_space-MNI152NLin2009cAsym_desc-preproc_T1w.nii.gz
sub-RevS01_ses-T1_acq-mprage_hemi-L_inflated.surf.gii                           sub-RevS01_ses-T1_acq-mprage_space-MNI152NLin2009cAsym_dseg.nii.gz
sub-RevS01_ses-T1_acq-mprage_hemi-L_midthickness.surf.gii                       sub-RevS01_ses-T1_acq-mprage_space-MNI152NLin2009cAsym_label-CSF_probseg.nii.gz
sub-RevS01_ses-T1_acq-mprage_hemi-L_pial.surf.gii                               sub-RevS01_ses-T1_acq-mprage_space-MNI152NLin2009cAsym_label-GM_probseg.nii.gz
sub-RevS01_ses-T1_acq-mprage_hemi-L_smoothwm.surf.gii                           sub-RevS01_ses-T1_acq-mprage_space-MNI152NLin2009cAsym_label-WM_probseg.nii.gz
sub-RevS01_ses-T1_acq-mprage_hemi-L_sulc.shape.gii

./fmap:
sub-RevS01_ses-T1_fmapid-auto00000_desc-coeff0_fieldmap.nii.gz     sub-RevS01_ses-T1_fmapid-auto00000_desc-preproc_fieldmap.json
sub-RevS01_ses-T1_fmapid-auto00000_desc-coeff1_fieldmap.nii.gz     sub-RevS01_ses-T1_fmapid-auto00000_desc-preproc_fieldmap.nii.gz
sub-RevS01_ses-T1_fmapid-auto00000_desc-magnitude_fieldmap.nii.gz

./func:
sub-RevS01_ses-T1_task-motor_boldref.nii.gz
sub-RevS01_ses-T1_task-motor_desc-confounds_timeseries.json
sub-RevS01_ses-T1_task-motor_desc-confounds_timeseries.tsv
sub-RevS01_ses-T1_task-motor_from-scanner_to-boldref_mode-image_xfm.txt
sub-RevS01_ses-T1_task-motor_from-scanner_to-T1w_mode-image_xfm.txt
sub-RevS01_ses-T1_task-motor_from-T1w_to-scanner_mode-image_xfm.txt
sub-RevS01_ses-T1_task-motor_space-MNI152NLin2009cAsym_boldref.nii.gz
sub-RevS01_ses-T1_task-motor_space-MNI152NLin2009cAsym_desc-aparcaseg_dseg.nii.gz
sub-RevS01_ses-T1_task-motor_space-MNI152NLin2009cAsym_desc-aseg_dseg.nii.gz
sub-RevS01_ses-T1_task-motor_space-MNI152NLin2009cAsym_desc-brain_mask.json
sub-RevS01_ses-T1_task-motor_space-MNI152NLin2009cAsym_desc-brain_mask.nii.gz
sub-RevS01_ses-T1_task-motor_space-MNI152NLin2009cAsym_desc-preproc_bold.json
sub-RevS01_ses-T1_task-motor_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz
sub-RevS01_ses-T1_task-rest_boldref.nii.gz
sub-RevS01_ses-T1_task-rest_desc-confounds_timeseries.json
sub-RevS01_ses-T1_task-rest_desc-confounds_timeseries.tsv
sub-RevS01_ses-T1_task-rest_from-scanner_to-boldref_mode-image_xfm.txt
sub-RevS01_ses-T1_task-rest_from-scanner_to-T1w_mode-image_xfm.txt
sub-RevS01_ses-T1_task-rest_from-T1w_to-scanner_mode-image_xfm.txt
sub-RevS01_ses-T1_task-rest_space-MNI152NLin2009cAsym_boldref.nii.gz
sub-RevS01_ses-T1_task-rest_space-MNI152NLin2009cAsym_desc-aparcaseg_dseg.nii.gz
sub-RevS01_ses-T1_task-rest_space-MNI152NLin2009cAsym_desc-aseg_dseg.nii.gz
sub-RevS01_ses-T1_task-rest_space-MNI152NLin2009cAsym_desc-brain_mask.json
sub-RevS01_ses-T1_task-rest_space-MNI152NLin2009cAsym_desc-brain_mask.nii.gz
sub-RevS01_ses-T1_task-rest_space-MNI152NLin2009cAsym_desc-preproc_bold.json
sub-RevS01_ses-T1_task-rest_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz
sub-RevS01_ses-T1_task-visual_boldref.nii.gz
sub-RevS01_ses-T1_task-visual_desc-confounds_timeseries.json
sub-RevS01_ses-T1_task-visual_desc-confounds_timeseries.tsv
sub-RevS01_ses-T1_task-visual_from-scanner_to-boldref_mode-image_xfm.txt
sub-RevS01_ses-T1_task-visual_from-scanner_to-T1w_mode-image_xfm.txt
sub-RevS01_ses-T1_task-visual_from-T1w_to-scanner_mode-image_xfm.txt
sub-RevS01_ses-T1_task-visual_space-MNI152NLin2009cAsym_boldref.nii.gz
sub-RevS01_ses-T1_task-visual_space-MNI152NLin2009cAsym_desc-aparcaseg_dseg.nii.gz
sub-RevS01_ses-T1_task-visual_space-MNI152NLin2009cAsym_desc-aseg_dseg.nii.gz
sub-RevS01_ses-T1_task-visual_space-MNI152NLin2009cAsym_desc-brain_mask.json
sub-RevS01_ses-T1_task-visual_space-MNI152NLin2009cAsym_desc-brain_mask.nii.gz
sub-RevS01_ses-T1_task-visual_space-MNI152NLin2009cAsym_desc-preproc_bold.json
sub-RevS01_ses-T1_task-visual_space-MNI152NLin2009cAsym_desc-preproc_bold.nii.gz

If you need more information, I will gladly provide some more.
Thank you for your time,

Antonin HUDRY

danlurie · October 14, 2025, 12:00am

I’m not familiar with fmristroke, but it looks like that error message is talking about missing BOLD derivatives. It doesn’t say what files exactly it is failing to find, but I don’t see anything critical missing from your fmriprep outputs, and the filenames for the derivatives should be identical regardless of whether cost function masking is (or is not) used during registration.

Do you still have your fmriprep working directory? One way to confirm that the lesion mask was used is to check the masks used during the norm workflow. If the cost function masking has been applied correctly, there should be a moving_mask image in the working directory for that workflow, and viewing the image should (depending on the case) show either the (native-space) T1 brain mask with the lesion voxels subtracted, or a global mask with lesion voxels subtracted.

I’d tell you where exactly to look but I don’t have a working directory from an appropriate fmriprep run to refer back to. Happy to help you figure out where to look if you want to paste your working directory structure.