Metadata-Version: 2.1
Name: bnnsurv
Version: 0.1.3
Summary: TensorFlow 2.x Bayesian Neural Network for Survival Analysis
Home-page: https://github.com/thecml/UE-BNNSurv
Author: Christian Marius Lillelund
Author-email: chr1000@gmail.com
License: UNKNOWN
Keywords: Deep Learning,Neural Network,Bayesian Learning,Survival Analysis
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.6
Classifier: Programming Language :: Python :: 3.7
Classifier: Programming Language :: Python :: 3.8
Classifier: Programming Language :: Python :: 3.9
Classifier: License :: OSI Approved :: MIT License
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Other Audience
Classifier: Intended Audience :: Science/Research
Classifier: Intended Audience :: Education
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Operating System :: OS Independent
Description-Content-Type: text/markdown
Requires-Dist: numpy (~=1.22)
Requires-Dist: scikit-survival (~=0.22)
Requires-Dist: tensorflow (>=2.11)
Requires-Dist: tensorflow-probability (>=0.19)

# Uncertainty Estimation in Deep Bayesian Survival Models

# *UPDATE 11/16/23: pip package now available. Use "pip install bnnsurv". Tested with TensorFlow 2.13 and TensorFlow Probability 0.21. See [test file](https://github.com/thecml/UE-BNNSurv/blob/main/tests/test_bnn_surv.py) for how to use.

This repository is the official TensorFlow implementation of [Uncertainty Estimation in Deep Bayesian Survival Models](https://ieeexplore.ieee.org/document/10313466), BHI 2023.

The proposed method is implemented based on [TensorFlow Probability](https://github.com/tensorflow/probability).

<b>Full paper is available on IEEE Xplore: https://ieeexplore.ieee.org/document/10313466</b>

<p align="left"><img src="https://github.com/thecml/UE-BNNSurv/blob/main/img/BNN.png" width="40%" height="40%">

In this work, we introduce the use of Bayesian inference techniques for survival analysis in neural networks that rely on the Coxâ€™s proportional hazard assumption, for which we discuss a new flexible and effective architecture. We implement three architectures: a fully-deterministic neural network that acts as a baseline, a Bayesian model using variational inference and one using Monte-Carlo Dropout.

Experiments show that the Bayesian models improve predictive performance over SOTA neural networks in a test dataset with few samples (WHAS500, 500 samples) and provide comparable performance in two larger ones (SEER and SUPPORT, 4024 and 8873 samples, respectively)

<p align="center"><img src="https://github.com/thecml/UE-BNNSurv/blob/main/img/seer_surv_all_models.png" width="30%" height="30%" /> <img src="https://github.com/thecml/UE-BNNSurv/blob/main/img/seer_surv_grade_mcd.png" width="30%" height="30%" /> <img src="https://github.com/thecml/UE-BNNSurv/blob/main/img/seer_surv_pdf.png" width="31%" height="31%" />


License
--------
To view the license for this work, visit https://github.com/thecml/UE-BNNSurv/blob/main/LICENSE


Requirements
----------------------
To run the models, please refer to [Requirements.txt](https://github.com/thecml/UE-BNNSurv/blob/main/requirements.txt).

Install auton-survival manually from Git:
```
pip install git+https://github.com/autonlab/auton-survival.git
```
Code was tested in virtual environment with `Python 3.8`, `TensorFlow 2.11` and `TensorFlow Probability 0.19`


Training
--------
- Make directories `mkdir results` and `mkdir models`.

- Please refer to `paths.py` to set appropriate paths. By default, results are in `results` and models in `models`

- Network configuration using best hyperparameters are found in `configs/*`

- Run the training code:

```
# SOTA models
python train_sota_models.py

# BNN Models
python train_bnn_models.py
```


Evaluation
--------
- After model training, view the results in the `results` folder.


Visualization
---------
- Run the visualization notebooks:
```
jupyter notebook plot_survival_curves.ipynb
jupyter notebook plot_survival_time.ipynb
```


Citation
--------
```
@inproceedings{lillelund_uncertainty_2023,
  author={Lillelund, Christian Marius and Magris, Martin and Pedersen, Christian Fischer},
  booktitle={2023 IEEE EMBS International Conference on Biomedical and Health Informatics (BHI)}, 
  title={Uncertainty Estimation in Deep Bayesian Survival Models}, 
  year={2023},
  pages={1-4},
  doi={10.1109/BHI58575.2023.10313466}
}
```


