Code for Multi-Task Deep Model with Margin Ranking Loss for Lung Nodule Analysis on IEEE Transactions on Medical Imaging (TMI).
This repository provides the PyTorch implementation for our TMI paper "Multi-Task Deep Model with Margin Ranking Loss for Lung Nodule Analysis". Our model can output a more robust benign-malignant classification result with persuasive semantic feature scores compared to other CAD techniques which can only output classification results.
Python == 2.7.13
PyTorch == 0.3.0
tensorboardX == 0.9
numpy == 1.14.3
Download and unzip this project:
git clone https://github.com/lihaoliu-cambridge/mtmr-net.git
cd mtmr-net
Download resnet50.pth into ./logs/middle_result_logs/imagenet/
folder.
Download the original LIDC-IDRI dataset into ./data/
folder
The preprocessing methods can be found in below two links:
https://github.com/zhwhong/lidc_nodule_detection
https://github.com/jcausey-astate/NoduleX_code
Finish you mode(training or test), and run it.
cd mtmr-net
python main.py
If you use our code for your research, please cite our paper:
@article{liu2019multi,
title={Multi-task deep model with margin ranking loss for lung nodule analysis},
author={Liu, Lihao and Dou, Qi and Chen, Hao and Qin, Jing and Heng, Pheng-Ann},
journal={IEEE transactions on medical imaging},
volume={39},
number={3},
pages={718--728},
year={2019},
publisher={IEEE}
}
Please open an issue or email lhliu1994@gmail.com for any questions.
:kissing_smiling_eyes:Thanks my dearest brother Yong for this beautiful figure.