kulbear/ivus-segmentation-icsm2018

IVUS Image Segmentation - ICSM 2018

By Ji Yang, Lin Tong, Mehdi Faraji and Anup Basu.

Multimedia Research Centre, Department of Computing Science, University of Alberta.

Introduction

This repository contains the original models and code described in the paper "IVUS-Net: An Intravascular Ultrasound Segmentation Network" (https://arxiv.org/abs/1806.03583).

Abstract

IntraVascular UltraSound (IVUS) is one of the most effective imaging modalities that provides assistance to experts in order to diagnose and treat cardiovascular diseases. We address a central problem in IVUS image analysis with Fully Convolutional Network (FCN): automatically delineate the lumen and media-adventitia borders in IVUS images, which is crucial to shorten the diagnosis process or benefits a faster and more accurate 3D reconstruction of the artery. Particularly, we propose an FCN architecture, called IVUS-Net, followed by a post-processing contour extraction step, in order to automatically segments the interior (lumen) and exterior (media-adventitia) regions of the human arteries. The proposed work, to the best of our knowledge, is the first deep learning based method for segmentation of both the lumen and the media vessel walls in 20MHz IVUS B-mode images that achieves the best results without any manual intervention.

Citation

@article{yang2018ivus,
	author = {Ji Yang and Lin Tong and Mehdi Faraji and Anup Basu},
	title = {IVUS-Net: An Intravascular Ultrasound Segmentation Network},
	journal = {arXiv preprint arXiv:1806.03583},
	year = {2018}
}

ivus-segmentation-icsm2018's Issues

Training loss value is very large

Hi，
I got your code and ran it. But the training loss value is always very large and it keeps 190 000 levels.

About your model of training

Hello:
I would like to borrow your training model to test some pictures. Since I don't have data set for training, could you please send me your training model if possible? Thank you very much.

关于IVUSchallenge2011的数据

你好：

在读到您的文章《IVUS-Net: An Intravascular Ultrasound Segmentation Network》时，发现您提供了相关的代码，想重复一下您的实验。但是原数据网址已经不能提供下载了，请问您有什么解决的办法吗

谢谢

"md_train" files

We are trying your program, but get the error message that it cannot find the file that starts with "md_train_". We would very much appreciate your help.

More detailed instructions on README.md

hi,
I would like to train your model on my dataset. I wonder if you could be more specific on your README.md, like eg, how to run your code step by step? It would be super nice to do that!

where is the ivus daaset?

Thank you for your code. I searched but not see the ivus data, plz tell us where do we download it? Thanks.

json 文件丢失

你好，我发现加载在工程里面并没有json配置文件,可以发一下吗，谢谢了