ISBI2018_PETCT_Segmentation

This repository contains the code (in TensorFlow) for "3D fully convolutional networks for co-segmentation of tumors on PET-CT images" paper (ISBI 2018). Compared to the previous semi-automated methods, this method is highly automated without manually user-defined seeds.

UPDATED

Uploaded the DFCN-CoSeg training and testing code for our extended work published in https://aapm.onlinelibrary.wiley.com/doi/abs/10.1002/mp.13331 (MP2018), which provided much details compared to the ISBI2018 paper.
Uploaded our previous trained models for CT-Only, PET-Only and DFCN-CoSeg networks studied in MP2018. The models can be downloaded in (1) BaiduYun (https://pan.baidu.com/s/1tCsjfuckkU9IH8O4xewsRQ Password: tfkt), or (2) https://app.box.com/s/9r7zxfcs5y9kr5woa1bze8v2lgz48ryv.
As for now, I cannot install the outdated tensorflow_gpu==1.4 in my working Ubuntu 20.04, so I uploaded two cases of PET-CT images and the testing code using tensorflow_gpu==2.3, interested readers can check the test.sh script. Please note that we just use the tensorflow_gpu==2.3 in the testing code, not for training.
With regarding to the PET SUV computation, please refer to the NCI-QIICR project (http://qiicr.org/tool/PETDICOM/), they have introduced an implementation as an extension for the open source 3D Slicer software (https://www.slicer.org/).

CT/PET Segmentation Results on One Patient

1. CT image

2. PET_SUV image

3. Ground Truth Segmentation on CT image

4. Ground Truth Segmentation on PET_SUV image

5. Prediction on CT image

6. Prediction on PET_SUV image

7. Wrong Predictions on CT image

8. Wrong Predictions on PET_SUV image

Dependencies

Python2.7
TensorFlow(1.4.0+)
DLTK
other libraries

Citation

If you find this useful, please cite our work as follows:

@INPROCEEDINGS{zszhong2018isbi_petct,
  author={Z. Zhong and Y. Kim and L. Zhou and K. Plichta and B. Allen and J. Buatti and X. Wu},
  booktitle={2018 IEEE 15th International Symposium on Biomedical Imaging (ISBI 2018)},
  title={3D fully convolutional networks for co-segmentation of tumors on PET-CT images},
  year={2018},
  volume={},
  number={},
  pages={228-231},
  keywords={Biomedical imaging;Computed tomography;Image segmentation;Lung;Three-dimensional displays;Tumors;co-segmentation;deep learning;fully convolutional networks;image segmentation;lung tumor segmentation},
  doi={10.1109/ISBI.2018.8363561},
  ISSN={},
  month={April},
}

@article{zszhong2018mp_petct,
  author = {Zhong, Zisha and Kim, Yusung and Plichta, Kristin and Allen, Bryan G. and Zhou, Leixin and Buatti, John and Wu, Xiaodong},
  title = {Simultaneous cosegmentation of tumors in PET-CT images using deep fully convolutional networks},
  journal = {Medical Physics},
  volume = {46},
  number = {2},
  pages = {619-633},
  keywords = {cosegmentation, deep learning, nonsmall cell lung cancer (NSCLC), tumor contouring},
  doi = {10.1002/mp.13331},
  url = {https://aapm.onlinelibrary.wiley.com/doi/abs/10.1002/mp.13331},
  eprint = {https://aapm.onlinelibrary.wiley.com/doi/pdf/10.1002/mp.13331},
  year = {2019}
}

Contacts

[email protected]

Any discussions or concerns are welcomed!

zhongzisha / isbi2018_petct_segmentation Goto Github PK

isbi2018_petct_segmentation's Introduction

ISBI2018_PETCT_Segmentation

CT/PET Segmentation Results on One Patient

1. CT image

2. PET_SUV image

3. Ground Truth Segmentation on CT image

4. Ground Truth Segmentation on PET_SUV image

5. Prediction on CT image

6. Prediction on PET_SUV image

7. Wrong Predictions on CT image

8. Wrong Predictions on PET_SUV image

Dependencies

Citation

Contacts

isbi2018_petct_segmentation's People

Contributors

Stargazers

Watchers

Forkers

isbi2018_petct_segmentation's Issues

Recommend Projects

Recommend Topics

Recommend Org