-
please refer to the implementation code: https://doi.org/10.5281/zenodo.6539438
-
Prepare your data.
-
Download the pre-train Res2Net parameters and put it into ./lib/res_weight/
-
https://drive.google.com/file/d/16_bh91WdaAKPFnwkMkDNMaQUTTIHboWi/view?usp=sharing
- run python train.py
- Download our pre-trained model's parameter then put it into your-own-path
- https://drive.google.com/file/d/12uqe20GdX3-DxeszPqdlIB2o8svyh_vs/view?usp=sharing
- run test.py
If you find our work useful or our work gives you any insights, please cite:
@article{meng2022dual,
title={Dual Consistency Enabled Weakly and Semi-Supervised Optic Disc and Cup Segmentation with Dual Adaptive Graph Convolutional Networks},
author={Meng, Yanda and Zhang, Hongrun and Zhao, Yitian and Gao, Dongxu and Hamill, Barbra and Patri, Godhuli and Peto, Tunde and Madhusudhan, Savita and Zheng, Yalin},
journal={IEEE Transactions on Medical Imaging},
year={2022},
publisher={IEEE}
}