Code for testing and reproduicing results for MRI local study project:
Modl and blinps code according to the following paper
Anish Lahiri, Guanhua Wang, Sai Ravishankar, Jeffrey A. Fessler, (2021). "Blind Primed Supervised (BLIPS) Learning for MR Image Reconstruction." IEEE Transactions on Medical Imaging http://doi.org/10.1109/TMI.2021.3093770; arXiv preprint arXiv:2104.05028.
The code is made up of three components: * single coil two channel study *,* multi coil two channel global study *, and *multi coil two channel local model*(with PyTorch > 1.7.0). * Additionally, we used BART to generate the dataset.
make_two_channel_dataset specifies and show how to make the two channel dataset based on the modification form https://github.com/JeffFessler/BLIPSrecon
global dataset specifies the data loader for MRI image loading from multicoil MR measurements for global case.
local network dataset and local netowrk dataset oracle specifies the data loader for MRI image loading from multicoil MR measurements for noise local case and oracle local case.
train local unet can be used for local model training and testing reconstruction from undersampled mulit-coil k-space measurements using UNet training.
transfer learning local network can be used for local model training and testing reconstruction from undersampled mulit-coil k-space measurements using DIDN training.