Welcome to PREDICT
===================================

The PREDICT project *(Pragmatic Recalibration and Evaluation of Drift In Clinical Tools)*
aims to improve the way that clinical prediction models are deployed and maintained.

Although this project is primarily focused on the UK healthcare system,
the tools and methods developed are applicable to any binary prediction model.

The PREDICT project is a research project funded by the 
National Insitute for Health and Care Research (NIHR206843) which aims to:

* Build software to help detect and repair temporal drift `Github`_
* Understand public opinion around the use of AI for clinical decision making, and deployment issues
* Raise the profile of this issue with relevant stakeholders, and suggest potential regulatory solutions

.. _Github: https://github.com/sdrelton/PREDICT

To contact the team, please email the project lead
Dr Samuel Relton (s.d.relton@leeds.ac.uk)

.. image:: images/temporal_drift_comic_strip_with_typed_text.png
   :width: 100%
   :align: center
   :alt: Comic strip showing the problem of temporal drift in clinical prediction models

Links
-----
* `Github code repository`_
* :ref:`Documentation <documentation>`
* :ref:`Newsletters on project progress <Newsletters>`

.. _Github code repository: https://github.com/sdrelton/PREDICT

Team
----
The core PREDICT team is:

* Samuel Relton (s.d.relton@leeds.ac.uk)
* Zoe Hancox (z.l.hancox@leeds.ac.uk)
* Kate Best (k.e.best@leeds.ac.uk)
* Oliver Todd (o.todd@leeds.ac.uk)
* Barbara Hartley


.. _documentation:

Documentation
=============
Here is the full documentation for the PREDICT software,
the associated Github with examples of usage is `here`_.

.. _here: https://github.com/sdrelton/PREDICT

.. toctree::
   :maxdepth: 2
   :caption: Contents:

   PREDICT
   Models
   Metrics
   Plots
   Tests

Indices and tables
------------------

* :ref:`genindex`
* :ref:`modindex`
* :ref:`search`