DeepReg is an open-source toolkit for research in medical image registration using deep learning.
The current version bases on TensorFlow 2. This toolkit contains implementations for unsupervised- and weaky-supervised algorithms with their combinations and variants, with a practical focus on diverse clinical applications, as in provided examples.
This is still under development. However, some functionalities can be accessed already.
Read the tutorials and documentation at DeepReg.