DeepReg

DeepReg is a freely available, community-supported open-source toolkitfor research and education in medical image registration using deeplearning.

The current version is implemented as aTensorFlow2-based framework,and contains implementations for unsupervised- and weakly-supervisedalgorithms with their combinations and variants. DeepReg has a practicalfocus on growing and diverse clinical applications, as seen in theprovided examples -DeepReg Demos.

Get involved and help make DeepReg better!

Features

DeepReg extends and simplifies workflows for medical imaging researchersworking in TensorFlow 2, and can be easily installed and used forefficient training and rapid deployment of deep-learning registrationalgorithms.

DeepReg is designed to be used with minimal programming or scripting,owing to its built-in command line tools.

Our development and all related work involved in the project is public,and released under the Apache 2.0 license.

Contact

For development matters, pleaseraise an issue.

For matters regarding theCode of Conduct, such as a complaint,please email the DeepReg Development Team:DeepRegNet@gmail.com.

Alternatively, please contact one or more members of the CoC Committee as appropriate: Nina Montana Brown (nina.brown.15@ucl.ac.uk), Ester Bonmati (e.bonmati@ucl.ac.uk), Matt Clarkson (m.clarkson@ucl.ac.uk).

WEISS LogoCME Logo

Contributors

DeepReg is maintained by a team of developers and researchers.People with significant contributions to DeepReg are acknowledged in theContributor List.

This open-source initiative started within University College London,with support from the Wellcome/EPSRC Centre for Interventional andSurgical Sciences (WEISS), and partial support from theWellcome/EPSRC Centre for Medical Engineering (CME).