Skip to content

Commit

Permalink
Merge branch 'master' of https://github.com/autonlab/auton-survival
Browse files Browse the repository at this point in the history
  • Loading branch information
@chiragnagpal
chiragnagpal committed Mar 15, 2022
2 parents 887aa42 + 45a5114 commit 490d570
Show file tree
Hide file tree
Showing 2 changed files with 192 additions and 118 deletions.
304 changes: 189 additions & 115 deletions README.md
View file
Original file line number Diff line number Diff line change
@@ -1,3 +1,4 @@

[![Build Status](https://travis-ci.org/autonlab/DeepSurvivalMachines.svg?branch=master)](https://travis-ci.org/autonlab/DeepSurvivalMachines)
   
[![codecov](https://codecov.io/gh/autonlab/DeepSurvivalMachines/branch/master/graph/badge.svg?token=FU1HB5O92D)](https://codecov.io/gh/autonlab/DeepSurvivalMachines)
Expand All @@ -6,196 +7,269 @@
   
[![GitHub Repo stars](https://img.shields.io/github/stars/autonlab/auton-survival?style=social)](https://github.com/autonlab/auton-survival)

<img align="right" width=50% src=https://ndownloader.figshare.com/files/34052981>

<img align=right style="align:right;" src="https://ndownloader.figshare.com/files/34052981" width=30%>

<br>

Package: `auton-survival`
-------------
Python package `auton_survival` provides a flexible API for various problems
in survival analysis, including regression, counterfactual estimation,
and phenotyping.

What is Survival Analysis?
--------------------------

Python package `dsm` provides an API to train the Deep Survival Machines
and associated models for problems in survival analysis. The underlying model
is implemented in `pytorch`.
**Survival Analysis** involves estimating when an event of interest, \( T \)
would take places given some features or covariates \( X \). In statistics
and ML these scenarious are modelled as regression to estimate the conditional
survival distribution, \( \mathbb{P}(T>t|X) \). As compared to typical
regression problems, Survival Analysis differs in two major ways:

For full documentation of the module, please see https://autonlab.github.io/auton-survival/
* The Event distribution, \( T \) has positive support ie.
\( T \in [0, \infty) \).
* There is presence of censoring ie. a large number of instances of data are
lost to follow up.

What is Survival Analysis?
------------------------
The Auton Survival Package
---------------------------

**Survival Analysis** involves estimating when an event of interest, _*T*_
would take place given some features or covariates _*X*_. In statistics
and ML, these scenarios are modelled as regression to estimate the conditional
survival distribution, _*P*_(_T_>t|_X_).
As compared to typical regression problems, Survival Analysis differs in two major ways:
The package `auton_survival` is repository of reusable utilities for projects
involving censored Time-to-Event Data. `auton_survival` allows rapid
experimentation including dataset preprocessing, regression, counterfactual
estimation, clustering and phenotyping and propnsity adjusted evaluation.

* The Event distribution, _*T*_ has positive support i.e. _*T*_ ∈ [0, ∞).
* There is presence of censoring i.e. a large number of instances of data are
lost to follow up.

Survival Regression
-------------------

Deep Survival Machines
----------------------
Currently supported Survival Models are:

### `auton_survival.models.dsm.DeepSurvivalMachines`
### `auton_survival.models.dcm.DeepCoxMixtures`
### `auton_survival.models.cph.DeepCoxPH`


### `auton_survival.estimators`

This module provids a wrapper to model survival datasets with standard
survival (time-to-event) analysis methods. The use of the wrapper allows
a simple standard interface for multiple different survival regression methods.

`auton_survival.estimators` also provides convenient wrappers around other popular
python survival analysis packages to experiment with the following
survival regression estimators

<img align="left" width=30% src=https://ndownloader.figshare.com/files/26367844> <img width=100% src=https://ndownloader.figshare.com/files/25259852>
- Random Survival Forests (`pysurvival`):
- Weibull Accelerated Failure Time (`lifelines`) :

### `auton_survival.experiments`

**Deep Survival Machines (DSM)** is a **fully parametric** approach to model
Time-to-Event outcomes in the presence of Censoring, first introduced in
[\[1\]](https://arxiv.org/abs/2003.01176).
In the context of Healthcare ML and Biostatistics, this is known as 'Survival
Analysis'. The key idea behind Deep Survival Machines is to model the
underlying event outcome distribution as a mixure of some fixed \( K \)
parametric distributions. The parameters of these mixture distributions as
well as the mixing weights are modelled using Neural Networks.
Modules to perform standard survival analysis experiments. This module
provides a top-level interface to run `auton_survival` style experiments
of survival analysis, involving cross-validation style experiments with
multiple different survival analysis models at different horizons of
event times.

#### Usage Example
The module further eases evaluation by automatically computing the
*censoring adjusted* estimates of the Metrics of interest, like
**Time Dependent Concordance Index** and **Brier Score** with **IPCW**
adjustment.

```python
from dsm import DeepSurvivalMachines
model = DeepSurvivalMachines()
model.fit()
model.predict_risk()
# auton_survival Style Cross Validation Experiment.
from auton_survival import datasets
features, outcomes = datasets.load_topcat()

from auton_survival.experiments import SurvivalCVRegressionExperiment

# instantiate an auton_survival Experiment by
# specifying the features and outcomes to use.
experiment = SurvivalCVRegressionExperiment(features, outcomes)

# Fit the `experiment` object with a Cox Model
experiment.fit(model='cph')

# Evaluate the performance at time=1 year horizon.
scores = experiment.evaluate(time=1.)

print(scores)
```

Recurrent Deep Survival Machines
--------------------------------
<img width=100% src=https://ndownloader.figshare.com/files/28329918>

Phenotyping and Knowledge Discovery
-----------------------------------

**Recurrent Deep Survival Machines (RDSM)** builds on the original **DSM**
model and allows for learning of representations of the input covariates using
**Recurrent Neural Networks** like **LSTMs, GRUs**. Deep Recurrent Survival
Machines is a natural fit to model problems where there are time dependendent
covariates.
### `auton_survival.phenotyping`

`auton_survival.phenotyping` allows extraction of latent clusters or subgroups
of patients that demonstrate similar outcomes. In the context of this package,
we refer to this task as **phenotyping**. `auton_survival.phenotyping` allows:

Deep Convolutional Survival Machines
------------------------------------
- **Unsupervised Phenotyping**: Involves first performing dimensionality
reduction on the inpute covariates \( x \) followed by the use of a clustering
algorithm on this representation.

Predictive maintenance and medical imaging sometimes requires to work with
image streams. Deep Convolutional Survival Machines extends **DSM** and
**DRSM** to learn representations of the input image data using
convolutional layers. If working with streaming data, the learnt
representations are then passed through an LSTM to model temporal dependencies
before determining the underlying survival distributions.
- **Factual Phenotyping**: Involves the use of structured latent variable
models, `auton_survival.models.dcm.DeepCoxMixtures` or
`auton_survival.models.dsm.DeepSurvivalMachines` to recover phenogroups that
demonstrate differential observed survival rates.

> :warning: **Not Implemented Yet!**
- **Counterfactual Phenotyping**: Involves learning phenotypes that demonstrate
heterogenous treatment effects. That is, the learnt phenogroups have differential
response to a specific intervention. Relies on the specially designed
`auton_survival.models.cmhe.DeepCoxMixturesHeterogenousEffects` latent variable model.

Deep Cox Mixtures
------------------
<img width=100% align="center" src=https://ndownloader.figshare.com/files/28316535>
Dataset Loading and Preprocessing
---------------------------------

The Cox Mixture involves the assumption that the survival function
of the individual to be a mixture of K Cox Models. Conditioned on each
subgroup Z=k; the PH assumptions are assumed to hold and the baseline
hazard rates is determined non-parametrically using an spline-interpolated
Breslow's estimator.
For full details on Deep Cox Mixture, refer to the paper:
Helper functions to load and prerocsss various time-to-event data like the
popular `SUPPORT`, `FRAMINGHAM` and `PBC` dataset for survival analysis.

<a href="https://proceedings.mlr.press/v149/nagpal21a">Deep Cox Mixtures
for Survival Regression. Machine Learning in Health Conference (2021)</a>

### `auton_survival.datasets`

Installation
------------
```python
# Load the SUPPORT Dataset
from auton_survival import dataset
features, outcomes = datasets.load_dataset('SUPPORT')
```

### `auton_survival.preprocessing`
This module provides a flexible API to perform imputation and data
normalization for downstream machine learning models. The module has
3 distinct classes, `Scaler`, `Imputer` and `Preprocessor`. The `Preprocessor`
class is a composite transform that does both Imputing ***and*** Scaling with
a single function call.

```python
# Preprocessing loaded Datasets
from auton_survival import datasets
features, outcomes = datasets.load_topcat()

from auton_survival.preprocessing import Preprocessing
features = Preprocessor().fit_transform(features,
cat_feats=['GENDER', 'ETHNICITY', 'SMOKE'],
num_feats=['height', 'weight'])

# The `cat_feats` and `num_feats` lists would contain all the categorical and
# numerical features in the dataset.

```console
foo@bar:~$ git clone https://github.com/autonlab/auton-survival.git
foo@bar:~$ cd auton-survival
foo@bar:~$ pip install -r requirements.txt
```

Examples
---------
Evaluation and Reporting
-------------------------

1. [Deep Survival Machines on the SUPPORT Dataset](https://nbviewer.jupyter.org/github/autonlab/DeepSurvivalMachines/blob/master/examples/DSM%20on%20SUPPORT%20Dataset.ipynb)
2. [Recurrent Deep Survival Machines on the PBC Dataset](https://nbviewer.org/github/autonlab/DeepSurvivalMachines/blob/master/examples/RDSM%20on%20PBC%20Dataset.ipynb)
### `auton_survival.metrics`

Helper functions to generate standard reports for common Survival Analysis tasks.

References
----------
Citing and References
----------------------

Please cite the following papers if you are using the `dsm` package.
Please cite the following papers if you are using the `auton_survival` package.

[1] [Deep Survival Machines:
Fully Parametric Survival Regression and
Representation Learning for Censored Data with Competing Risks.
IEEE Journal of Biomedical \& Health Informatics (2021)](https://arxiv.org/abs/2003.01176)</a>
Representation Learning for Censored Data with Competing Risks."
IEEE Journal of Biomedical and Health Informatics (2021)](https://arxiv.org/abs/2003.01176)</a>

```
@article{nagpal2021deep,
title={Deep Survival Machines: Fully Parametric Survival Regression and\
Representation Learning for Censored Data with Competing Risks},
@article{nagpal2021dsm,
title={Deep survival machines: Fully parametric survival regression and representation learning for censored data with competing risks},
author={Nagpal, Chirag and Li, Xinyu and Dubrawski, Artur},
journal={IEEE Journal of Biomedical and Health Informatics},
year={2021}
volume={25},
number={8},
pages={3163--3175},
year={2021},
publisher={IEEE}
}
```


[2] [Deep Parametric Time-to-Event Regression with Time-Varying Covariates.
AAAI Spring Symposium (2021)](http://proceedings.mlr.press/v146/nagpal21a.html)</a>
[2] [Deep Parametric Time-to-Event Regression with Time-Varying Covariates. AAAI
Spring Symposium (2021)](http://proceedings.mlr.press/v146/nagpal21a.html)</a>

```
@InProceedings{pmlr-v146-nagpal21a,
title = {Deep Parametric Time-to-Event Regression with Time-Varying Covariates},
author = {Nagpal, Chirag and Jeanselme, Vincent and Dubrawski, Artur},
booktitle = {Proceedings of AAAI Spring Symposium on Survival Prediction - Algorithms, Challenges, and Applications 2021},
series = {Proceedings of Machine Learning Research},
publisher = {PMLR},
@InProceedings{pmlr-v146-nagpal21a,
title={Deep Parametric Time-to-Event Regression with Time-Varying Covariates},
author={Nagpal, Chirag and Jeanselme, Vincent and Dubrawski, Artur},
booktitle={Proceedings of AAAI Spring Symposium on Survival Prediction - Algorithms, Challenges, and Applications 2021},
series={Proceedings of Machine Learning Research},
publisher={PMLR},
}
```

[3] [Deep Cox Mixtures for Survival Regression. Machine Learning for Healthcare (2021)](https://proceedings.mlr.press/v149/nagpal21a)
[3] [Deep Cox Mixtures for Survival Regression. Conference on Machine Learning for
Healthcare (2021)](https://arxiv.org/abs/2101.06536)</a>

```
@InProceedings{nagpal2021dcm,
title={Deep Cox Mixtures for Survival Regression},
@inproceedings{nagpal2021dcm,
title={Deep Cox mixtures for survival regression},
author={Nagpal, Chirag and Yadlowsky, Steve and Rostamzadeh, Negar and Heller, Katherine},
booktitle={Proceedings of the 6th Machine Learning for Healthcare Conference},
booktitle={Machine Learning for Healthcare Conference},
pages={674--708},
year={2021},
volume={149},
series={Proceedings of Machine Learning Research},
publisher={PMLR},
}
organization={PMLR}
}
```

[4] [Counterfactual Phenotyping with Censored Time-to-Events (2022)](https://arxiv.org/abs/2202.11089)</a>

```
@article{nagpal2022counterfactual,
title={Counterfactual Phenotyping with Censored Time-to-Events},
author={Nagpal, Chirag and Goswami, Mononito and Dufendach, Keith and Dubrawski, Artur},
journal={arXiv preprint arXiv:2202.11089},
year={2022}
}
```

## Installation

```console
foo@bar:~$ git clone https://github.com/autonlab/auton_survival
foo@bar:~$ pip install -r requirements.txt
```

Compatibility
-------------
`dsm` requires `python` 3.5+ and `pytorch` 1.1+.
`auton_survival` requires `python` 3.5+ and `pytorch` 1.1+.

To evaluate performance using standard metrics
`dsm` requires `scikit-survival`.
`auton_survival` requires `scikit-survival`.

Contributing
------------
`dsm` is [on GitHub]. Bug reports and pull requests are welcome.
`auton_survival` is [on GitHub]. Bug reports and pull requests are welcome.

[on GitHub]: https://github.com/chiragnagpal/deepsurvivalmachines
[on GitHub]: https://github.com/autonlab/auton-survival

License
-------
MIT License

Copyright (c) 2020 Carnegie Mellon University, [Auton Lab](http://www.autonlab.org)
Copyright (c) 2022 Carnegie Mellon University, [Auton Lab](http://autonlab.org)

Permission is hereby granted, free of charge, to any person obtaining a copy of this
software and associated documentation files (the "Software"), to deal in the Software
without restriction, including without limitation the rights to use, copy, modify,
merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit
persons to whom the Software is furnished to do so, subject to the following conditions:
Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all copies
or substantial portions of the Software.
The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED,
INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR
PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE
FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR
OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
DEALINGS IN THE SOFTWARE.
THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.

<img align="right" height ="120px" src="https://www.cs.cmu.edu/~chiragn/cmu_logo.jpeg">
<img align="right" height ="110px" src="https://www.cs.cmu.edu/~chiragn/auton_logo.png">

<br><br><br><br><br>
<br><br><br><br><br>
Loading

0 comments on commit 490d570

Please sign in to comment.