search

yeatmanlab / pyAFQ

Automated Fiber Quantification ... in Python
http://yeatmanlab.github.io/pyAFQ/
BSD 2-Clause "Simplified" License
57 stars 33 forks source link

Use fiber-to-bundle coherence for cleaning #607

Closed arokem closed 7 months ago

arokem commented 3 years ago

FBC is a slightly more elaborate way of cleaning bundles (see: https://dipy.org/documentation/1.3.0./examples_built/fiber_to_bundle_coherence/). I wonder whether it improves cleaning generally, relative to what we are currently doing. We should give it a try.

jyeatman commented 3 years ago

Would love to see it! The 3D Gaussian of sea can definitely be improved upon

On Sun, Nov 22, 2020 at 12:19 PM Ariel Rokem notifications@github.com wrote:

FBC is a slightly more elaborate way of cleaning bundles (see: https://dipy.org/documentation/1.3.0./examples_built/fiber_to_bundle_coherence/). I wonder whether it improves cleaning generally, relative to what we are currently doing. We should give it a try.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/yeatmanlab/pyAFQ/issues/607, or unsubscribe https://github.com/notifications/unsubscribe-auth/AANRRSSW5HZ4QDB5VZW5ODLSRFW5HANCNFSM4T6WMLMQ .

-- Sent from a mobile device. Please excuse typos

36000 commented 7 months ago

This is a neat method which provides good and reliable results. However, it seems to take a lot longer to run than the method we currently use, so I don't think its useful to us. Also, I cannot think of a case where it would succeed and our multi-round mahalanobis distance cleaning would fail. So, if someone wants to make a PR to add this to pyAFQ, it would not take long and I welcome it. But, I do not think this is a priority, so I am going to close this issue.