Dynamic-weighting Hierarchical Segmentation Network for Medical Images

by Xiaoqing Guo.

Summary:

Intoduction:

This repository is for our MedIA paper "Dynamic-weighting Hierarchical Segmentation Network for Medical Images"

Framework:

Usage:

Requirement:

Pytorch 1.3 Python 3.6

Preprocessing:

Clone the repository:

git clone https://github.com/CityU-AIM-Group/DW-HieraSeg.git
cd DW-HieraSeg 
bash sh_hierasegCVC.sh
bash sh_dw_hierasegCVC.sh
bash sh_hierasegISIC.sh
bash sh_dw_hierasegISIC.sh

Data preparation:

Dataset should be put into the folder './data'. For example, if the name of dataset is CVC, then the path of dataset should be './data/CVC/', and the folder structure is as following.

ThresholdNet
|-data
|--CVC
|---images
|---labels
|---train.txt
|---test.txt
|---valid.txt

The content of 'train.txt', 'test.txt' and 'valid.txt' should be just like:

26.png
27.png
28.png
...

Pretrained model:

You should download the pretrained model from Google Drive, and then put it in the './model' folder for initialization.

Well trained model:

You could download the trained model from Google Drive, which achieves 82.328% in Jaccard score on the EndoScene testing dataset. Put the model in directory './models'.

Citation:

@article{guo2021dynamic,
  title={Dynamic-weighting Hierarchical Segmentation Network for Medical Images},
  author={Guo, Xiaoqing and Yang, Chen, Yuan, Yixuan},
  journal={Medical Image Analysis},
  year={2021}
}

Questions:

Please contact "xiaoqingguo1128@gmail.com"