AthenaEPI / dmipy

The open source toolbox for reproducible diffusion MRI-based microstructure estimation
MIT License
96 stars 30 forks source link

How to generate fiber bundles #126

Open cxmDREAM opened 1 month ago

cxmDREAM commented 1 month ago

hello,I am very interested in generating fiber bundles and I would like to use NODDI's Watson to generate fiber bundles, Dear, do you have an example of how to use FODS combined with DIPY to generate fiber bundles

rutgerfick commented 1 month ago

Hello!

So in the NODDI example we show how to visualize the FODs from the fitted model representation using dipy https://nbviewer.org/github/AthenaEPI/dmipy/blob/master/examples/example_noddi_watson.ipynb.

However, it's likely that you'll need to change the dipy code a little since they probably updated some things in their visualization code. It seems this tutorial of theirs shows how to visualize the FODs for another model, so you can just merge the 2 examples :-). Happy to see you're using dmipy! https://workshop.dipy.org/documentation/1.7.0/examples_built/07_reconstruction/reconst_shore/#sphx-glr-examples-built-07-reconstruction-reconst-shore-py

cxmDREAM commented 1 month ago

Hello @rutgerfick , I think I may not have made myself clear. I would like to know how to generate fiber bundle tracking based on the Watson model in Dimpy, Do you have any similar examples? The effect is like this
https://docs.dipy.org/stable/examples_built/visualization/viz_roi_contour.html#

rutgerfick commented 1 month ago

Hello, So it seems that the fiber tracking example uses csa_peaks which appears to be the fiber orientation distribution of the CSA model, generated at a spherical tesselation defined by default_sphere.

If my understanding is correct, then all you need to do is replace csa_peaks by fods from the noddi_watson example, as long as you generate the FODs at the same tesselation. Does that work?