0.3.0
0.3.0 reflects renewed maintenance for ASLPrep.
The primary focuses of this release are:
(1) fixing easily-fixable bugs,
(2) disabling broken features with difficult-to-fix bugs,
(3) cleaning up the codebase,
(4) expanding test coverage,
and (5) ensuring that outputs are BIDS-compliant.
To that end, we have unfortunately had to temporarily drop support for multi-PostLabelingDelay data,
as well as PASL data with the Q2TIPS BolusCutOffTechnique.
We will work on fixing these features for the next release.
Additionally, this release includes a number of breaking changes.
We have renamed several of the outputs to ensure that they are BIDS-compliant.
These outputs may need to change again in the future,
but that will happen in the next minor release (0.4.0) at the earliest.
See the table below for a list of the changed filenames.
| Description | <0.3.0 | 0.3.0 |
|---|---|---|
| CBF time series | _cbf.nii.gz |
_desc-timeseries_cbf.nii.gz |
| Mean CBF | _mean_cbf.nii.gz |
_cbf.nii.gz |
| CBF time series after SCORE denoising | _desc-score_cbf.nii.gz |
_desc-scoreTimeseries_cbf.nii.gz |
| Mean CBF after SCORE denoising | _desc-score_mean_cbf.nii.gz |
_desc-score_cbf.nii.gz |
| Bolus arrival time/arterial transit time | _desc-bat_cbf.nii.gz |
_att.nii.gz |
Additionally, we have changed the atlases that are used for the parcellated CBF files.
We have brought the atlases in line with those used by XCP-D.
However, in the near future, we plan to update these atlases again, so that they are synchronized across
ASLPrep, XCP-D, and QSIPrep, so please be aware of that upcoming change.
What's Changed
🛠 Breaking Changes
- Rename BIDS-noncompliant derivatives by @tsalo in #216
- Use correct bolus values in CBF calculation and temporarily disable Q2TIPS, multi-PLD, and no-BolusCutOff processing by @tsalo in #235
- Disable SCORE/SCRUB for GE/short runs and get GE workflow working by @tsalo in #248
- Parallelize parcellation and replace atlases by @tsalo in #254
- Remove "mean" from mean CBF filenames and add "timeseries" to 4D CBF filenames by @tsalo in #257
🎉 Exciting New Features
- Hardcode T1blood for some field strengths and use Zhang 2013 formula for others by @tsalo in #243
- Estimate labeling efficiency based on ASL type and number of background suppression pulses by @tsalo in #244
🐛 Bug Fixes
- Pin looseversion by @tsalo in #211
- Correct ordering of probseg maps by @josephmje in #192
- Convert inversion times to a list for compatility with BASIL interface by @tsalo in #222
- Pin networkx version by @tsalo in #227
- Use LabelingEfficiency as BASIL
--alphaparameter by @tsalo in #233 - Remove BIDS-noncompliant m0z and cbf files by @tsalo in #236
- Replace RepetitionTime with RepetitionTimePreparation by @tsalo in #245
- Index aslcontext rows with "cbf", not "CBF" by @tsalo in #252
Other Changes
- Use aslprep_build as base Docker image - attempt 2 by @tsalo in #204
- Autoformat with black and isort by @tsalo in #205
- Replace relative imports with absolute ones by @tsalo in #206
- Remove unused functions and classes by @tsalo in #207
- Rename PEP8-noncompliant classes by @tsalo in #208
- Consistently use f-strings for string formatting by @tsalo in #209
- Work on fixing docstrings and splitting modules by @tsalo in #210
- Remove unused arguments throughout package by @tsalo in #212
- Replace smriprep calls with dependency calls by @tsalo in #217
- Replace pybids calls with dependency calls by @tsalo in #213
- Use custom formatting for workflow connections by @tsalo in #220
- Improve documentation by @tsalo in #218
- Add step to run unit tests by @tsalo in #221
- Collect associated files at beginning of workflow by @tsalo in #246
- Refactor confounds and QC metrics by @tsalo in #256
New Contributors
- @tsalo made their first contribution in #204
- @josephmje made their first contribution in #192
Full Changelog: 0.2.8...0.3.0
