search

nipreps / mriqc

Automated Quality Control and visual reports for Quality Assessment of structural (T1w, T2w) and functional MRI of the brain
http://mriqc.readthedocs.io
Apache License 2.0
284 stars 128 forks source link

Head motion estimation on DWI is poor #1283

Open oesteban opened 2 months ago

oesteban commented 2 months ago

We're using 3dVolreg (from AFNI), copying the procedures from BOLD counterparts. In the long term, I think education should do this better than anything else, but in the short term, we may want to talk with Paul Taylor and find a good configuration of AFNI for DWI.

arokem commented 2 months ago

In the long term, I think education should do this better than anything else

Do you mean "eddymotion"? I'd think that would still be pretty time-consuming to run in most case, beyond what you'd want here?

Might also want to try dipy.align.motion_correction

oesteban commented 2 months ago

Do you mean "eddymotion"?

hehehe yes :D

I'd think that would still be pretty time-consuming to run in most case, beyond what you'd want here?

I think we want to get it in good shape and have simple models that are competitively fast.

Might also want to try dipy.align.motion_correction

Sure! I'd be happy to try it.