GD-Net

This is the official source code for the paper "GCN-Enhanced Spatial-Spectral Dual-Encoder Network for Simultaneous Segmentation of Retinal Layers and Fluid in OCT Images"
Coded by Zhilin Zhou

Citation

If you use this code for any academic purpose, please cite: link to the paper

title={GCN-Enhanced Spatial-Spectral Dual-Encoder Network for Simultaneous Segmentation of Retinal Layers and Fluid in OCT Images}
author={Guogang Cao, Zhilin Zhou, Yan Wu, Zeyu Peng, Rugang Yan, Yunqing Zhang, Bin Jiang}
journal={Biomedical Signal Processing and Control}
year={2024}
organization={Elsevier}

Environment

The requirements.txt file includes the required libraries for this project.

conda create --name GDNet python=3.8.17
conda activate GDNet
pip install -r requirements.txt

Dataset

The dataset folder hierarchy should look like this:

dataset
--DUKE
----train
------img
------mask
----test
------img
------mask
----eval
------img
------mask

1.DUKE DME: https://people.duke.edu/~sf59/Chiu_BOE_2014_dataset.htm
2.RETOUCH: https://retouch.grand-challenge.org/
3.Peripapillary OCT dataset: http://www.yuyeling.com/project/mgu-net/

Train and test

Run the following script to train and test our model

python train&test.py --name GDNet --batch-size 1 --epoch 50 --lr 0.001

Acknowledgements

The codes are built on Li. We sincerely appreciate the authors for sharing their codes.

Contact

If you have any questions,please contact 226142168@mail.sit.edu.cn