TaG-Net：Topology-aware Graph Network for Centerline-based Vessel Labeling

This is the official PyTorch implementation for the TaG-Net method to handle the head and neck vessel labeling based on CTA image.

• Publication: Yao et al. TaG-Net: Topology-aware Graph Network for Centerline-based Vessel Labeling. IEEE Transactions on Medical Imaging, 2023.
• Citation:

@ARTICLE{10032183,
  author={Yao, Linlin and Shi, Feng and Wang, Sheng and Zhang, Xiao and Xue, Zhong and Cao, Xiaohuan and Zhan, Yiqiang and Chen, Lizhou and Chen, Yuntian and Song, Bin and Wang, Qian and Shen, Dinggang},
  journal={IEEE Transactions on Medical Imaging}, 
  title={TaG-Net: Topology-Aware Graph Network for Centerline-Based Vessel Labeling}, 
  year={2023},
  volume={42},
  number={11},
  pages={3155-3166},
  doi={10.1109/TMI.2023.3240825}}

Abstract

We propose a novel framework for centerline-based vessel labeling. The framework contains two separate models (SegNet and TaG-Net). SegNet is utilized to offer the initial vessel segmentation. TaG-Net is used for centerline labeling. Besides, a graph-based vessel completion method is proposed and utilized in test stage to alleviate the vessel interruption and adhesion resulted from the initial vessel segmentation. Experimental results show that our proposed method can significantly improve both head and neck vessel segmentation and labeling performance.

Framework

SegNet

nnU-Net (3D U-Net cascade) is trained on our dataset to offer the initial vessel segmentation.

Hu range is set as [0, 800] (Window width/level = 800/400).

TaG-Net

Vessel Completion

Adhesion Removal

Usage: Preparation

Environment

• Ubuntu 18.04
• Python 3.7 (recommend Anaconda3)
• Pytorch 0.4.1
• CMake >= 3.10.2
• CUDA 9.0 + cuDNN 7.1

Installation

Clone

git clone https://github.com/PRESENT-Y/TaG-Net.git 
cd TaG-Net

Pytorch 0.4.1

conda create -n TaG-Net python=3.7
conda activate TaG-Net
conda install pytorch==0.4.1 torchvision cuda90 -c pytorch

Other Dependencies (e.g., dgl, networkx, mayavi and dijkstra3d)

pip install -r requirements.txt 

Build

mkdir build && cd build
cmake .. && make  

Data Preparation

Download

• We have provided sample data for testing.
• Sample data, the corresponding ground truth, and our result can be downloaded at Google Drive.
• Download and put them in ./SampleData.

Centerline Vascular Graph Construction

• Generate centerline from initial segmentation mask.
• Transform centerline image into point set.
• Construct centerline vascular graph from point set.
• Remove isolated nodes and triangles.

python ./GraphConstruction/gen_cl_graph.py

For visualization of the centerline graph, you can run the following python files.

python ./GraphConstruction/vis_cl_graph.py

Usage: Training

CUDA_VISIBLE_DEVICES=0 python ./train.py

You can modify ./cfgs/config_train.yaml.

Usage: Evaluation

CUDA_VISIBLE_DEVICES=0 python ./test.py

Usage: Vessel Completion

We conduct the vessel completion based on the labeled vascular graph (output of the TaG-Net).

sh ./VesselCompletion/vessel_completion.sh

For visualization of the labeled centerline graph, you can run the following python files.

python ./VesselCompletion/vis_labeled_cl_graph.py

License

The code is released under GPL License (see LICENSE file for details).

Acknowledgements

• This code repository refers to nnUNet, pointnet.pytorch, PointNet2_PyTorch, and Relation-Shape-CNN.
• We use Mayavi for point set and centerline vascular graph visualization.
• We use EvaluateSegmentation for computing metrics.
• We thank all contributors for their awesome and efficient code bases.

Contact

If you have some ideas or questions about our research, please contact [email protected].