Summary of what happened:
Hi,
We have run fMRIPrep on a dataset with 7 BOLD runs and 3 field maps. For some of the participants, in the “Alignment between the anatomical reference of the field map and the target EPI” section of the output html, one image, at far x coordinates, is blank - all other images are present (we are unable to share the correct images due to it being a patient cohort). E.g., for one participant the blank image is at x=-44 and for another participant it is at x=-37.
When comparing the BOLD nii and the field map nii in FSL at these coordinates, we are still within the brain mask. In addition, in other figures in the html output, the images at x=-44 are present (e.g., Brain mask and (anatomical/temporal) CompCor ROIs).
Is this an issue with alignment between BOLD and field maps or perhaps just an issue with the html generation.
Any help would be greatly appreciated!
Command used (and if a helper script was used, a link to the helper script or the command generated):
PASTE CODE HERE
Version:
PUT VERSION HERE
Environment (Docker, Singularity / Apptainer, custom installation):
PUT ENVIRONMENT HERE
Data formatted according to a validatable standard? Please provide the output of the validator:
PASTE VALIDATOR OUTPUT HERE
Relevant log outputs (up to 20 lines):
PASTE LOG OUTPUT HERE
Screenshots / relevant information:
Hi @MRI_New,
In the future please use the Software Support post category and post template. You can see I edited in for you this time.
As long as the images look fine in an image viewer at those coordinates you are fine.
Best,
Steven
Hi Steven,
Thank you for your reply! Apologies for not correctly using the template for the post.
As a follow-up, we have also experienced an issue within the B0 field mapping: Preprocessed mapping of phase-difference acquisition section of the html. Where the green outline is not capturing the entire brain. As you can see in this example here:
Hi @MRI_New,
Please add in the missing information requested by the post template so we can better understand your issue.
Best,
Steven
Command used (and if a helper script was used, a link to the helper script or the command generated):
We have run fMRIPrep on 75 participants who have 7 BOLD runs (6 task, 1 rest) and 3 field maps. For about 5% we have seen in the html output that the green line in B0 field mapping: Preprocessed mapping of phase-difference acquisition section does not cover the whole brain. As evidenced in the screenshot below.
PASTE CODE HERE
participant \
--output-spaces MNI152NLin2009cAsym \
--dummy-scans 5 \
--write-graph \
--nthreads 8 --omp-nthreads 4 \
--stop-on-first-crash \
--fs-no-reconall \
Version:
fMRIPrep 24.1.0
Environment (Docker, Singularity / Apptainer, custom installation):
Apptainer
Data formatted according to a validatable standard? Please provide the output of the validator:
BIDS validated
Relevant log outputs (up to 20 lines):
PASTE LOG OUTPUT HERE
Screenshots / relevant information:
Can you try with FreeSurfer enabled and use the most recent version? Are the fieldmaps associated in an appropriate way (that is, the fieldmaps are correcting the BOLD scans they are closest to in time)? Any fieldmap/T1w/BOLD quality concerns for those subjects?
Hi Steven,
This is data that was collected 12-13 years ago and we are trying to replicate the preprocessing pipeline that was conducted in SPM, now in fMRIPrep.
In the original processing they assigned each of the field maps to the two fMRI runs that occurred directly prior to the field map and so we have replicated that with our IntendedFor
i.e.,
MPRAGE
Run1
Run2
FieldMap 1 (assigned to run 1 &2)
Run3
Run4
FieldMap 1 (assigned to run 3 &4)
Run5
Run6
FieldMap 1 (assigned to run 5 &6)
Due to the data being conducted a while ago, the quality is appropriate for then.
We are not using freesurfer as we are not running any surface-based analysis and it is very computationally expensive
Hi @MRI_New,
You can try SynBOLD-DisCo, it may work better than legacy fieldmaps.
Disabling FreeSurfer also results in worse anatomical registration due to boundary based registration being disabled, regardless if you are doing any surface analyses.
Best,
Steven