-
Notifications
You must be signed in to change notification settings - Fork 296
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
ENH: Add MP-PCA/NORDIC denoising through dwidenoise #3395
base: master
Are you sure you want to change the base?
Conversation
Codecov ReportAttention: Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #3395 +/- ##
==========================================
- Coverage 72.06% 71.07% -0.99%
==========================================
Files 57 59 +2
Lines 4231 4464 +233
Branches 455 482 +27
==========================================
+ Hits 3049 3173 +124
- Misses 1067 1156 +89
- Partials 115 135 +20 ☔ View full report in Codecov by Sentry. |
It is intended to address this at some stage; for tracking see MRtrix3/mrtrix3#2274 and MRtrix3/mrtrix3#3035.
I've added a note to MRtrix3/mrtrix3#3035.
Agree 100%; we're invested locally in being able to modulate downstream applications appropriately based on this. It would be good to:
There's a few branches in parallel that have proposed changes to
That could be done. It would be reasonably easy to add an export option to
When I've done this myself, I've concatenated data across echoes into the fourth dimension, denoised, and then separated the volumes back out into echoes. That is I think more principled given that the noise level should be identical. It's not super difficult to do. You may however want to clamp the maximal patch size: once (# echoes x # volumes) exceeds 343 the default changes to a 9x9x9 patch and that runs very slowly. This behaviour should be better with MRtrix3/mrtrix3#2742 / MRtrix3/mrtrix3#3029, which won't have such step changes in patch size. Some other points to consider:
|
Thanks @Lestropie for looping me in this conversation.
|
Posted as MRtrix3/mrtrix3#3046. Would need implementation within the underlying denoising command; whether fMRIPrep would want to utilise that is beyond my control.
It might be possible to mitigate this if the full noise map profile is known a priori, and voxels can be normalised to unit variance upon insertion into the Casorati matrix? MRtrix3/mrtrix3#3041
Have more flexible patch size in this implementation with spherical kernels: MRtrix3/mrtrix3#2742
Cool; it would be useful for me to know from someone more familiar with the mathematics / simulations that it's sensible. |
Closes #3308.
Potential blockers
-rank
option to output a DOF map, but that feature hasn't been released yet. See ENH dwidenoise: Optionally output degrees of freedom map MRtrix3/mrtrix3#2312.Changes proposed in this pull request
init_bold_native_wf
.--thermal-denoise-method
to specify which approach to use, if any.