Deep learning for fully-automated prediction of overall survival in patients with oropharyngeal cancer using FDG PET imaging: an international retrospective study. Clinical Cancer Research, OnlineFirst version was published on May 4, 2021
This is an example of running inference with trained DeepPET-OPSCC prognostic models. For the DeepPET-OPSCC-T model, the inputs are the PET (SUV) ROI image and corresponding tumor mask; for the DeepPET-OPSCC-TN model, an additional distance map is needed. An illustrative example of the input data is provided in the sample_data folder. The default data format is NIfTI (.nii.gz). If you want to try your own data, please first convert PET to image with SUV values and put in the 'ori_data' folder, the ROI/mask/dist map will automatically be saved in the folder 'sample_data'.
This inference code is based on Pytorch and has been tested with the latest version (1.8.0)
- Ananoconda environment installation and activation
conda create -n deeppet-py3 pip python=3.6
conda activate deeppet-py3
- Install Pytorch and required packages
conda install pytorch torchvision torchaudio cudatoolkit=11.0 -c pytorch -c conda-forge
pip install SimpleITK
pip install tqdm
pip install vtk
pip install scikit-image
- Compile NVIDIA-Kaolin for generating dist map, first download all files and then compile it with the following Command
python setup.py develop
- Run the inference code (trained models needed, availabe upon request)
python inference_DeepPETOPSCC.py
DeepPET-OPSCC trained prognostic models are available for research-use only. This tool is provided for research purposes only and no responsibility is accepted for clinical decisions arising from its use. COMMERCIAL USE IS PROHIBITED for the time being. For further information please email yaojiawen076@paii-labs.com.
If you find this repository useful in your research, please cite:
@article{Cheng2021,
title={Deep learning-based fully-automated prediction of overall survival in patients with oropharyngeal cancer using FDG PET imaging},
author={Nai-Ming Cheng, Jiawen Yao, Jinzheng Cai, Xianghua Ye, Shilin Zhao, Kui Zhao, Wenlan Zhou, Isabella Nogues, Yuankai Huo, Chun-Ta Liao, Hung-Ming Wang, Chien-Yu Lin, Li-Yu Lee, Jing Xiao, Le Lu, Ling Zhang, and Tzu-Chen Yen},
journal={Clinical Cancer Research}
}