Skip to content

Deep Cascade of Convolutional Neural Networks for MR Image Reconstruction: Implementation & Demo

License

Notifications You must be signed in to change notification settings

EmonLu/Deep-MRI-Reconstruction

This branch is up to date with js3611/Deep-MRI-Reconstruction:master.

Folders and files

NameName
Last commit message
Last commit date

Latest commit

d8a40ef · Jan 21, 2019

History

16 Commits
Nov 10, 2017
Jan 21, 2019
Oct 2, 2017
Jan 21, 2019
Jan 21, 2019
May 30, 2017
May 30, 2017
Jan 21, 2019
Nov 10, 2017
Oct 2, 2017
Jan 21, 2019
Jan 21, 2019

Repository files navigation

Deep Cascade of Convolutional Neural Networks and Convolutioanl Recurrent Nerual Networks for MR Image Reconstruction


Reconstruct MR images from its undersampled measurements using Deep Cascade of Convolutional Neural Networks (DC-CNN) and Convolutional Recurrent Neural Networks (CRNN-MRI). This repository contains the implementation of DC-CNN using Theano and Lasagne, and CRNN-MRI using PyTorch, along with simple demos. Note that the library requires the dev version of Lasagne and Theano, as well as pygpu backend for using CUFFT Library. PyTorch version needs to be higher than Torch 0.4. Some of the toy dataset borrowed from <http://mridata.org>.

1. 2D Reconstruction

Usage:

python main_2d.py --num_epoch 5 --batch_size 2

2. Dynamic Reconstruction

Reconstruct dynamic MR images from its undersampled measurements using DC-CNN with Data Sharing layer. Note that the library requires CUDNN in addition to the requirement specified above.

Usage:

python main_3d.py --acceleration_factor 4

3. Dynamic Reconstruction using Convolutional Recurrent Neural Networks

Reconstruct dynamic MR images from its undersampled measurements using Convolutional Recurrent Neural Networks. This is a pytorch implementation requiring Torch 0.4.

Usage:

python main_crnn.py --acceleration_factor 4

Citation and Acknowledgement

If you use the code for your work, or if you found the code useful, please cite the following works.

2D Reconstruction:

Schlemper, J., Caballero, J., Hajnal, J. V., Price, A., & Rueckert, D. A Deep Cascade of Convolutional Neural Networks for MR Image Reconstruction. Information Processing in Medical Imaging (IPMI), 2017

The paper is also available on arXiv: <https://arxiv.org/pdf/1703.00555.pdf>

Dynamic Reconstruction:

Schlemper, J., Caballero, J., Hajnal, J. V., Price, A., & Rueckert, D. A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstruction. ArXiv 1704.02422

The paper is also available on arXiv: <https://arxiv.org/pdf/1704.02422.pdf>

Dynamic Reconstruction using CRNN:

Qin, C., Schlemper, J., Caballero, J., Hajnal, J. V., Price, A., & Rueckert, D. Convolutional Recurrent Neural Networks for Dynamic MR Image Reconstruction. IEEE transactions on medical imaging (2018).

The paper is also available on arXiv: <https://arxiv.org/pdf/1712.01751.pdf>

About

Deep Cascade of Convolutional Neural Networks for MR Image Reconstruction: Implementation & Demo

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%