DeepReg is a freely available, community-supported open-source toolkit for research and education in medical image registration using deep learning.
- TensorFlow2-based for efficient training and rapid deployment;
- Implementing major unsupervised and weakly-supervised algorithms, with their combinations and variants;
- Focusing on growing and diverse clinical applications, with all DeepReg Demos using open-accessible data;
- Simple built-in command line tools requiring minimal programming and scripting;
- Open, permissible and research-and-education-driven, under the Apache 2.0 license.
DeepRegbeta will be released in Autumn 2020 - however, many tutorials, demos and much of the core functionality are already accessible.
- Documentation, tutorials and a quick start guide
- DeepReg Demos
- Code: https://github.com/DeepRegNet/DeepReg
- Issue tracker: https://github.com/DeepRegNet/DeepReg/issues
- Website: http://deepreg.net/
Get involved, and help make DeepReg better!
For guidance on making a contribution to DeepReg, see the contribution guidelines.
If you identified a registration application with openly accessible data, please consider contributing a DeepReg Demo.
The MICCAI Educational Challenge notebook can be accessed here or run it on Colab.