Matt Cieslak

@barisgenc12 I don't think environment variables in your shell session will have any effect on the qsiprep run inside of docker. Where did you change the path?

@36000 would you mind taking a look?

Can you try re-running it with `--skip-odf-reports`? There's something going on with the odf plotting in version 0.15+ that is causing unpredictable and silent errors

@WilliamFCB could you try pennbbl/qsiprep:unstable? This has the fix you proposed and should solve this issue

LPS+ is used because it is the coordinate system used by ITK/ANTs. FSL actually uses LAS+ (RPI) orientation internally. There is also a recent reconstruction workflow that will convert your...

ah, it's python 3.8 now so the docs need some updating

Hi Chandana, FSL does not use the image axes for bvecs *unless the image is in LAS+ voxel orientation*. DIPY uses the image axis orientation regardless of the voxel direction....

Hi! It looks like there are somehow no bvals or bvecs associated with your dwi file. Was the data preprocessed by QSIPrep?

Do both sequences have enough b>0 images for eddy to run? Also, you can keep both AP and PA scans in the dwi/ directory and use the `--separate-all-dwis` to have...

yes, it will look in the dwi files to see if any have b=0 images that are suitable for distortion correction. Don't bother putting anything in the fmap/ directory