a-fODFs recursive net

Code for paper "Unsupervised Asymmetric Fiber Orientation Distribution Functions Estimation on diffusion MRI".

[Notice]

The paper is submitted to MICCAI 2025, and partial codes are coming soon...

Setup

The main environment is:

• cuda 11.3
• torch 1.11.0
• nibabel 4.0.2
• PyGSP 0.5.1 (pip install git+https://github.com/epfl-lts2/pygsp.git@39a0665f637191152605911cf209fc16a36e5ae9#egg=PyGSP)
• dipy 1.7.0
• scipy 1.8.0

Datasets

You need to create a folder containing the dmri data. The file name in this folder must be:

1. dmri image: dmri_data.nii.gz
2. bvals file: bvals.bvals
3. bvecs file: bvecs.bvecs

Train Model and Inference on your data

bash estimate.sh