BFRnet

Xuanyu Zhu, Yang Gao, Feng Liu, Stuart Crozier, Hongfu Sun, 2022. BFRnet: A deep learning-based MR background field removal method for QSM of the brain containing significant pathological susceptibility sources

Setup BFRnet for SEPIA

  1. Download deepMRI from GitHub

  2. Download the pre-trained BFRnet here as mentioned in the instruction on GitHub

  3. Specify the full path to deepMRI code as ‘deepMRI_HOME’ in setup_BFRnet_environment.m in SEPIA_HOME/addons/bfr/BFRnet/

  4. Specify the full path to the pre-trained network (should be checkpoints/BFRnet_L2_64PS_24BS_45Epo_NewHCmix.mat) from (2) as ‘checkpoints’ in setup_BFRnet_environment.m

Your setup_BFRnet_environment.m should look something like this:

../../_images/BFRnet_setup.png

Warning

The support this method is still in an early stage and only tested on a Linux machine.

BFRnet panel

There is no algorithm parameter needed to be adjusted with this tool at the moment.

../../_images/BFRnet.png