Hello fellow MRtrix3 users,
I was wondering if anyone has experience in generating orientation dispersion entropy as mentioned by Enrico Kaden et al in 2016.
I’ve read through the post in MRtrix3 discussion: Measure of fibre orientation dispersion in MRTrix3
This given me an idea to start by using:
- sh2amp on wmfod.mif to obtain amplitudes from SH on uniform sphere
- ensure amplitudes are positive for near zero / negative numbers (mrcalc amp.mif 1e-10 -max)
- normalize amplitude image by dividing it with its sum
- mrcalc for applying: sum over N (log(norm_amp x N) x norm_amp)
The output I got for this makes sense to me in the cerebellum but not so much in the cerebrum.
I would think the cortical grey matter would also be around the same intensity as the cerebellar grey matter but it is bright, even as bright as the corpus callosum or internal capsule so I’m leaning on that this is artifactual.
Just a background for the data used: 79 directions, multishell b=0, 1500, 3000.
The wmfod.mif was created by dwi2fod msmt and normalized using mtnormalise.
Any advice / past-experience regarding this would be grateful!
Thank you
Danny