sgy1664 / ct_mar_attention

Codes and data for paper "Rigid and Non-rigid Motion Artifact Reduction in X-ray CT using Attention Module"

Rigid and Non-rigid Motion Artifacts Reduction in X-ray CT using Attention Module

This repository offers the data and code introduced in the following paper:

"Rigid and Non-rigid Motion Artifacts Reduction in X-ray CT using Attention Module".

Dependencies

To utilize our code, make sure you have installed Python and Tensorflow on your system.

In our experiments, python2.7 and tensorflow1.8 have been used.

Preparation

To download our code, clone this repository as follow:

git clone https://github.com/youngjun-ko/ct_mar_attention
cd ct_mar_attention

Prepare your data in following form:

Folder name: './data'
File format: '.npy' (numpy array)
Data format: 'NHWC'

• Our dataset and pre-trained models can be downloaded from here

• Pre-trained VGG model can be downloaded from here

Please note that the code and model have been updated on 2021.06.10.

Usage

To train and test our network, run:

python train.py
python test.py

Citation

If you find our work useful, please cite us:

@article{ko2021rigid,
  title={Rigid and non-rigid motion artifact reduction in X-ray CT using attention module},
  author={Ko, Youngjun and Moon, Seunghyuk and Baek, Jongduk and Shim, Hyunjung},
  journal={Medical Image Analysis},
  volume={67},
  pages={101883},
  year={2021},
  publisher={Elsevier}
}

Contact

E-mail: [e-mail]