DeepMRI: Exploiting deep residual network for fast parallel MR imaging with complex convolution
By Shanshan Wang, Huitao Cheng, Taohui Xiao, Leslie Ying, Ziwen Ke, Hairong Zheng, Dong Liang*
This repository contains the implementation of DeepMRI for parallel MR imaging using Tensorflow-gpu. Test code for Matlab and test data are also included. For a quick test, the trained weights can be used. The folder saved_model_vn contains the model of the trained VN dataset for the knee images.
The architecture of DeepMRI for parallel MR imaging
Reconstruction Results by classical SPIRiT/L1-SPIRiT and the proposed method
More results can be found in the paper.
Dependencies
- Python3
- Tensorflow
- numpy scipy