FDU-VTS / MMAC

The winning solutions from Team fdvts_mm in the MICCAI 2023 Myopic Maculopathy Analysis Challenge.

MMAC

This repo covers the winning solutions from Team fdvts_mm in the MICCAI 2023 Myopic Maculopathy Analysis Challenge.

Dataset

We download the datasets from MMAC2023 and DDR.

Task 1. Classification of Myopic Maculopathy

cd MMAC_task1

train a ResNet50 model

python main.py --challenge 1 --model resnet50 --visname resnet50

Task 2. Segmentation of Myopic Maculopathy Plus Lesions

cd MMAC_task2

train segmentation model

bash train.sh

Ensemble appoaches in MMAC_task2/ensemble_model.py

Reference

If you use this code, please cite the following papers:

@inproceedings{hou2023towards,
  title={Towards Label-Efficient Deep Learning for Myopic Maculopathy Classification},
  author={Hou, Junlin and Xu, Jilan and Xiao, Fan and Zhang, Bo and Xu, Yiqian and Zhang, Yuejie and Zou, Haidong and Feng, Rui},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  pages={31--45},
  year={2023},
  organization={Springer}
}

@inproceedings{xiao2023ensemble,
  title={Ensemble Deep Learning Approaches for Myopic Maculopathy Plus Lesions Segmentation},
  author={Xiao, Fan and Hou, Junlin and Xu, Jilan and Xu, Yiqian and Zhang, Bo and Zhang, Yuejie and Zou, Haidong and Feng, Rui},
  booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
  pages={46--55},
  year={2023},
  organization={Springer}
}