This repository is an unofficial pytorch implementation of DeepSurv: Personalized Treatment Recommender System Using A Cox Proportional Hazards Deep Neural Network. We reimplement the experiments in the paper, and the detailed understanding is available on my Blog.
- Pytorch>=0.4.0
- CPU or GPU
- Other packages can be installed with the following instruction:
pip install requirements.txt
Running the code with the following command.
python main.py
Note: You can modify some parameters in configs/*.ini to get your own specific models.
Simulated Linear | Simulated Nonlinear | WHAS | SUPPORT | METABRIC | Simulated Treatment | Rotterdam & GBSG | |
---|---|---|---|---|---|---|---|
Paper | 0.774019 | 0.648902 | 0.862620 | 0.618308 | 0.643374 | 0.582774 | 0.668402 |
Our implements | 0.778722 | 0.500512 | 0.844973 | 0.617398 | 0.637163 | 0.559965 | 0.670171 |
@article{Katzman2016DeepSurv,
title={DeepSurv: Personalized Treatment Recommender System Using A Cox Proportional Hazards Deep Neural Network},
author={Katzman, Jared and Shaham, Uri and Bates, Jonathan and Cloninger, Alexander and Jiang, Tingting and Kluger, Yuval},
journal={Bmc Medical Research Methodology},
volume={18},
number={1},
pages={24},
year={2016},
}