fix: Detect and apply precomputed fieldmaps #3439
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #3439 +/- ##
==========================================
- Coverage 72.99% 72.37% -0.63%
==========================================
Files 57 57
Lines 4369 4430 +61
Branches 548 570 +22
==========================================
+ Hits 3189 3206 +17
- Misses 1060 1107 +47
+ Partials 120 117 -3 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
effigies
force-pushed
the
fix/transforms
branch
from
1a8df30 to
eb1bb73
Compare
Member
Author
|
@mgxd A review would be appreciated. I've run this locally on ds000217 and its OpenNeuro derivatives (selected due to having a fieldmap and minimal derivatives). |
mgxd
reviewed
Mar 20, 2025
Collaborator
mgxd
left a comment
mgxd
left a comment
There was a problem hiding this comment.
few comments, but overall happy with this!
Member
Author
mgxd
reviewed
Mar 20, 2025
Co-authored-by: Mathias Goncalves <goncalves.mathias@gmail.com>
mgxd
approved these changes
Mar 20, 2025
effigies
commented
Mar 20, 2025
effigies
added a commit
that referenced
this pull request
Mar 25, 2025
25.0.0 (March 25, 2025)
New feature release in the 25.0.x series.
This release substantially improves support for pre-computed derivatives.
Previous releases would miss some derivatives and rerun the computations.
Note that derivatives from previous versions will be accepted,
so it should not be necessary to recompute derivatives from previous versions.
The recommended command line is::
fmriprep BIDS_DIR OUT_DIR participant --derivatives fmriprep=PRECOMP_DIR
Note that multiple derivatives can be specified, for example::
fmriprep BIDS_DIR OUT_DIR participant \
anat=PRECOMPUTED_ANATOMICAL_DIR \
func=PRECOMPUTED_FUNCTIONAL_DIR
When the same file is found in multiple derivatives, the last one found takes precedence.
Additionally, `--force-*` flags have been consolidated into a single
`--force` flag that can take multiple, space-separated arguments.
Structural processing changes
-----------------------------
We now output white, pial and midthickness fsLR meshes on the subject surface.
Look for `sub-<subject>_hemi-<L|R>_space-fsLR_*_<surf>.surf.gii` files.
Brain extraction has been modified slightly to more closely match the
`antsBrainExtraction.sh` workflow distributed by ANTs.
The impact should be minimal, but in rare cases this fixes a crash.
Fieldmap processing changes
---------------------------
SyN-SDC fieldmap filtering is now single-level, following the improvements
for gradient-echo fieldmaps in 24.1.
Jacobian-weighting during fieldmap unwarping is now on by default *only*
for PEPolar fieldmaps.
To enable for other fieldmap types, use `--force fieldmap-jacobian`.
All merged pull requests
------------------------
* FIX: Detect and apply precomputed fieldmaps (#3439)
* FIX: Calculate bold mask and dummy scans in transform-only runs (#3428)
* FIX: Use consistent skull-stripping pre- and post- SDC (#3415)
* FIX: Use removeprefix instead of lstrip or ternary operator (#3409)
* FIX: Listify sessions when generating reports (#3408)
* FIX: Ensure fieldmap is resampled correctly in report (#3387)
* FIX: Stop excluding FS minc_modify_header used during fallback registration (#3372)
* FIX: Repair and test query for precalculated baseline/boldref files (#3370)
* FIX: Repair search for precomputed transforms (#3369)
* ENH: Enable Jacobians only for PEPOLAR by default, allow forcing (#3443)
* ENH: Create `--force` flag that accepts a list, replacing individual `--force-*` flags (#3442)
* ENH: Output fsLR meshes on subject surfaces (#3411)
* ENH: Flexibilize "sophisticated" pepolar to allow monomodal execution (#3393)
* ENH: Update FSL packages for reported bug fixes (#3374)
* RF: Calculate RMSD from motion transforms (#3427)
* RF: Reconstruct motion confounds from minimal derivatives (#3424)
* RF: Replace deprecated pkgutil.find_loader (#3384)
* RF: Upgrade nitransforms and remove workarounds (#3378)
* DOC: Fix xfm extension in the outputs docs (#3435)
* DOC: Mention fMRIPost-AROMA in parser documentation (#3356)
* MNT: Remove CLI flags with expired deprecation periods (#3445)
* MNT: Update pinned environment (#3440)
* MNT: Bump pins, update RTD config (#3425)
* MNT: Declare linux/amd64 platform during Docker build (#3422)
* MNT: Bump astral-sh/setup-uv from 4 to 5 (#3417)
* MNT: Test support for Python 3.13 (#3416)
* MNT: Install Workbench CLI via conda (#3410)
* MNT: Update minimum dependencies, test with tox-uv (#3412)
* MNT: Install c3d through conda (#3382)
* CI: Fetch tags and 200 commits to support describe (#3381)
* CI: Build docker images in GHA, store cache inline and push to GHCR (#3380)
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is the last step for using pre-computed derivatives without recomputing. I'm able to target native, T1w, MNI and fsLR spaces using this.
This is mainly about fieldmaps in
fmriprep.workflows.base, but other things fixed in passing:--use-syn-sdcstill needs to be declared if SyN-SDC fieldmaps are to be used.There's still a minor bug, or missed opportunity, in
workflows.bold.fitto make sure that fieldmaps are reported even if coregistration is already done. It doesn't feel like it's worth holding this PR up to make that happen.