Releases: PennLINC/xcp_d
0.4.0rc1
What's Changed
🛠 Breaking Changes
- Change default highpass filter cutoff from 0.009 to 0.01 by @kahinimehta in #755
- Refactor anatomical workflow by @tsalo in #706
- Do not require
--combineruns
to generate DCAN QC files by @tsalo in #753 - Change QC filenames and fix
xcp_d-combineqc
command-line interface by @tsalo in #762 - Retain cohort entity in derivative filenames by @tsalo in #769
- Require the
--cifti
flag in order to use--warp-surfaces-native2std
by @tsalo in #770 - Zero out parcels with <50% coverage by @tsalo in #757
- Use constant padding and maximum padlen for temporal filtering by @tsalo in #779
🎉 Exciting New Features
- Add option to estimate brain radius from brain mask by @tsalo in #754
- Output warped atlases to derivatives by @tsalo in #647
- Add
min_coverage
parameter to threshold parcel coverage by @tsalo in #782 - Improve executive summary carpet plots by @tsalo in #747
- Output T2w images if available by @tsalo in #648
🐛 Bug Fixes
- Replace missing vertices' values with NaNs by @tsalo in #743
- Select MNI152NLin6Asym target space for T1w from CIFTI derivatives by @tsalo in #759
- Only generate brainsprite figures when the
--dcan_qc
flag is used by @tsalo in #766
Other Changes
- Add tests for cifti smoothness by @kahinimehta in #659
- Fix CIFTI downcasting test by @tsalo in #752
- Use pytest to test command-line interface by @tsalo in #740
- Upload coverage reports to CodeCov by @tsalo in #758
- Link to
xcp_d-examples
in documentation by @tsalo in #761 - Add information about preprocessing requirements by @tsalo in #772
- Replace MultiLabel interpolation with GenericLabel by @tsalo in #786
- Mention cosine regressors in aCompCor boilerplate by @tsalo in #788
- Update base Docker image by @tsalo in #799
Full Changelog: 0.3.2...0.4.0rc1
0.3.3rc4
Full Changelog: 0.3.3rc3...0.3.3rc4
0.3.3rc3
0.3.3rc2
0.3.3rc1
0.3.2
This release adopts a workbench show-scene-based brainsprite for the executive summary. It also removes the brainsprite figure from the nipreps report.
What's Changed
🎉 Exciting New Features
Other Changes
- Remove example data and fix workflow graphs by @tsalo in #738
- Replace CiftiDespike with connected nodes by @tsalo in #737
Full Changelog: 0.3.1...0.3.2
0.3.1
What's Changed
🛠 Breaking Changes
🎉 Exciting New Features
- Add
--dummy-scans
parameter and deprecate--dummytime
by @tsalo in #616 - Add
--bids-filter-file
parameter by @tsalo in #686 - Enable non-aggressive denoising with signal regressors by @tsalo in #697
- Improve identification of FreeSurfer derivatives by @tsalo in #719
- Collect preprocessed surfaces in new function by @tsalo in #731
🐛 Bug Fixes
- Remove dummy volumes from custom confounds files by @tsalo in #660
- Remove dummy volumes from beginning of each run in concatenation workflow by @tsalo in #661
- Retain noise components instead of dropping them in load_aroma by @tsalo in #670
- Use input_type to determine order of preferred spaces by @tsalo in #688
- Infer volumetric space from transform in executive summary with cifti data by @tsalo in #689
- Change input type from HPC to HCP by @kahinimehta in #691
- Replace failing non-aggressive AROMA denoising with working aggressive denoising by @tsalo in #693
- Select best volumetric space for transforms even when using
--cifti
flag by @tsalo in #695 - Downcast >32-bit files to 32-bit by @tsalo in #666
- Use appropriate intent codes for cifti outputs by @tsalo in #690
- Only concatenate processed runs by @tsalo in #713
- Do not use downcasted files as name sources by @tsalo in #712
- Warp segmentation file to appropriate space for carpet plots by @tsalo in #727
- Use brain mask in NIFTI connectivity workflow by @tsalo in #733
Other Changes
- Consolidate confounds at beginning of denoising workflows by @tsalo in #664
- Remove unused outputnodes from nifti and cifti workflows by @tsalo in #667
- Move QC/censoring plots into new workflow by @tsalo in #668
- Testing affines don't change across XCP runs by @kahinimehta in #649
- Load confounds via Nilearn functionality by @kahinimehta in #675
- Update docs by @kahinimehta in #682
- Add masks to package data by @kahinimehta in #715
- Use ds001491 fMRIPrep derivatives for tests by @tsalo in #698
- Add tests for confound loading function by @tsalo in #730
- Refactor executive summary workflow by @tsalo in #721
- Track start of workflow with sentry by @tsalo in #732
Full Changelog: 0.3.0...0.3.1
0.3.0
We are happy to announce a new minor release, with several backwards-incompatible changes.
Two big breaking changes are (1) there is a new --dcan-qc
flag that determines if the executive summary and DCAN-format QC files will be generated, and (2) custom confounds should now have headers, should be tab-delimited, and should have the same names as the fMRIPrep confounds, for easier indexing.
What's Changed
🛠 Breaking Changes
- Output ReHo as a CIFTI by @tsalo in #601
- Add
--dcan-qc
flag by @tsalo in #650 - Support custom confounds with headers by @tsalo in #642
🎉 Exciting New Features
- Use BIDSLayout in concatenation code by @tsalo in #600
- [ENH] Support Nibabies ingression into XCP by @kahinimehta in #602
- Add static ALFF and ReHo plots to report by @kahinimehta in #604
- Write confounds out to derivatives by @kahinimehta in #634
- Add column names to confounds df by @kahinimehta in #641
🐛 Bug Fixes
- Select BOLD files in a single space by @tsalo in #603
- Censor data in executive summary plots correctly by @kahinimehta in #614
- Only generate ALFF derivatives if bandpass filtering is enabled by @tsalo in #628
- Do not merge in concatenated files when re-running concatenation workflow by @tsalo in #633
- Map abbreviated entities to full names for templateflow get call by @tsalo in #654
- Ensure
dummyscans
is an integer in PlotSVGData by @tsalo in #655
Other Changes
- Use Nilearn for brainsprite generation by @tsalo in #607
- Cache the downloaded test data by @tsalo in #629
- Use BIDSLayout instead of globbing functions to collect necessary files by @tsalo in #621
- Start to standardize interface parameter calls by @tsalo in #638
- Simplify transform-getting functions by @tsalo in #623
- Lint with black without linting workflow connections by @tsalo in #640
Full Changelog: 0.2.0...0.3.0
0.2.2
This is a patch release for the 0.2 series. The main bug being fixed is that using --dummytime
was causing crashes in the executive summary workflow.
What's Changed
🎉 Exciting New Features
- Write confounds out to derivatives by @kahinimehta in #634
- Add column names to confounds df by @kahinimehta in #641
🐛 Bug Fixes
- Do not merge in concatenated files when re-running concatenation workflow by @tsalo in #633
- Map abbreviated entities to full names for templateflow get call by @tsalo in #654
- Ensure
dummyscans
is an integer in PlotSVGData by @tsalo in #655
Other Changes
- Cache the downloaded test data by @tsalo in #629
- Use BIDSLayout instead of globbing functions to collect necessary files by @tsalo in #621
- Start to standardize interface parameter calls by @tsalo in #638
- Simplify transform-getting functions by @tsalo in #623
- Lint with black without linting workflow connections by @tsalo in #640
Full Changelog: 0.2.1...0.2.2
0.2.1
This is a patch release for 0.2.0.
There is a known bug with the concatenation workflow, so we advise users not to use the -m
/--combineruns
option with this release.
What's Changed
🛠 Breaking Changes
🎉 Exciting New Features
- Use BIDSLayout in concatenation code by @tsalo in #600
- [ENH] Support Nibabies ingression into XCP by @kahinimehta in #602
- Add static ALFF and ReHo plots to report by @kahinimehta in #604
🐛 Bug Fixes
- Select BOLD files in a single space by @tsalo in #603
- Censor data in executive summary plots correctly by @kahinimehta in #614
- Only generate ALFF derivatives if bandpass filtering is enabled by @tsalo in #628
Other Changes
Full Changelog: 0.2.0...0.2.1