Skip to content

iMED-Lab/CS-Net

Repository files navigation

CS-Net: Channel and Spatial Attention Network for Curvilinear Structure Segmentation

Implementation of CS-Net: Channel and Spatial Attention Network for Curvilinear Structure Segmentation

For the details of 3D extended version of CS-Net, please refer to CS2-Net: Deep Learning Segmentation of Curvilinear Structures in Medical Imaging


Overview

The main contribution of this work is the publication of two scarce datasets in the medical image field. Plesae click the link below to access the details and source data.

Requirements

The attention module was implemented based on DANet. The difference between the proposed module and the original block is that we added a new 1x3 and 3x1 kernel convolution layer into spatial attention module. Plese refer to the paper for details.

Get Started

Using the train.py and predict.py to train and test the model on your own dataset, respectively.

Examples

  • Vessel segmentation on Fundus

  • Vessel segmentation on OCT-A images

  • Nerve fiber tracing on CCM

Citation

@inproceedings{mou2019cs,
title={CS-Net: channel and spatial attention network for curvilinear structure segmentation},
author={Mou, Lei and Zhao, Yitian and Chen, Li and Cheng, Jun and Gu, Zaiwang and Hao, Huaying and Qi, Hong and Zheng, Yalin and Frangi, Alejandro and Liu, Jiang},
booktitle={International Conference on Medical Image Computing and Computer-Assisted Intervention},
pages={721--730},
year={2019},
organization={Springer}
}

Useful Links

DRIVE http://www.isi.uu.nl/Research/Databases/DRIVE/
STARE http://www.ces.clemson.edu/ahoover/stare/
IOSTAR http://www.retinacheck.org/
ToF MIDAS http://insight-journal.org/midas/community/view/21
Synthetic https://github.com/giesekow/deepvesselnet/wiki/Datasets
VascuSynth http://vascusynth.cs.sfu.ca/Data.html

Releases

No releases published

Packages

No packages published

Languages