MIR.models.HyperMorph

HyperMorph

Original code retrieved from: https://github.com/voxelmorph/voxelmorph

Original paper:

Hoopes, Andrew, et al. “Hypermorph: Amortized hyperparameter learning for image registration.” Information Processing in Medical Imaging: 27th International Conference, IPMI 2021, Virtual Event, June 28–June 30, 2021, Proceedings 27. Springer International Publishing, 2021.

Balakrishnan, G., Zhao, A., Sabuncu, M. R., Guttag, J., & Dalca, A. V. (2019). VoxelMorph: a learning framework for deformable medical image registration. IEEE transactions on medical imaging, 38(8), 1788-1800.

Modified and tested by: Junyu Chen jchen245@jhmi.edu Johns Hopkins University

Classes

ConvBlock(*args, **kwargs)

Convolutional block with hypernetwork-generated weights.

ConvBlock_(*args, **kwargs)

Convolutional block with hyperparameter-conditioned bias.

CustomConv(*args, **kwargs)

Convolution with weights predicted by a hypernetwork.

HyperBlocks(*args, **kwargs)

MLP that embeds hyperparameters for hypernetwork conditioning.

HyperVxmDense(*args, **kwargs)

HyperMorph variant of VoxelMorph with hyperparameter conditioning.

Unet(*args, **kwargs)

U-Net backbone with hyperparameter conditioning.