Rigid registration for 3D MRI
The code is currently in development.
Data Augmentation
Rotate 3D image: /ext/image
3D-UNet
A 3D-UNet featured with unsampling and dowsampling layer is used to learn the shift flow between two different images.
CNN outputs three 3D flows.
Spatial Transformer
A 3D-spatial transformer, which is similar to the standard STN is utilized to do the registration task.
Trying to solve an optimization problem, which will get an affine transform matrix for spatial transformer to do the registration.
Citation
Based on medical image processing library Voxelmorph
Unsupervised Learning for Fast Probabilistic Diffeomorphic Registration
Adrian V. Dalca, Guha Balakrishnan, John Guttag, Mert R. Sabuncu
MICCAI 2018. eprint arXiv:1805.04605
An Unsupervised Learning Model for Deformable Medical Image Registration
Guha Balakrishnan, Amy Zhao, Mert R. Sabuncu, John Guttag, Adrian V. Dalca
CVPR 2018. eprint arXiv:1802.02604