Revision of dMRI artifact correction workflows #903
Conversation
|
I'm submitting this PR just in case I reach the deadline of the next release (please do not merge yet), still a lot of work going on. Also tests need be revised. |
When merged the new dipy interfaces I've written, this example should be updated adding a denoising node.
* ApplyTopup is now called with correct arguments when the reversed encoding image is only one b0 * Added documentation * Deprecated old workflows in nipype.workflows.dmri.fsl (added warnings in both documentation and code) * Updated CHANGES
Conflicts: CHANGES
Uses TOPUP and Eddy for correction. Tested on the new dmri_preprocess.py example with the fsl course data.
oesteban
changed the title
|
Changes Unknown when pulling 55457a2 on oesteban:enh/NewEPIArtifactCorrections into * on nipy:master*. |
|
@satra, @chrisfilo I need two text files to be passed as arguments to FLIRT. I added them under worflows/data and created an Other than this, I think these new workflows are ready to be merged. There is one new example based on the FSL course data that works correctly. I'm testing the remaining workflows. There are some stuff to refine (e.g. these two files, I opened a thread in the FSL forum trying to improve them). I will see how to write some real tests (as oppposed to doctests) using the fsl course data like those for bedpostx. |
|
This looks good to me. The only problem I have is the new naming convention you are proposing. I am leaning towards your approach - grouping workflows by modality rather than software packages. We need to be consistent though. Maybe it would be better to keep the old nomenclature in this PR and start a discussion/new PR about new convention? |
|
Ok, no problem on removing the headers, they are automatically added by my editor. Regarding the nomenclature, you mean I should move this code under fsl? No problem about that, I understand we first need to agree on the new nomenclature. |
|
@chrisfilo :
|
A lot of revisions under this PR. It'd be interesting to have it included into the next release.
Moved workflows fromnipype.workflows.dmri.fsltonipype.workflows.dmri.preprocessfor two reasons: 1) unlike Interfaces, workflows are not oriented to tools and therefore they might benefit from using several tools; 2) preprocess is a better name, as I intend to include resampling and denoising as in (http://nipy.org/dipy/examples_built/reslice_datasets.html#example-reslice-datasets, http://nipy.org/dipy/examples_built/denoise_nlmeans.html#example-denoise-nlmeans).