mridc.collections.reconstruction.models.cascadenet package

Submodules

mridc.collections.reconstruction.models.cascadenet.ccnn_block module

class mridc.collections.reconstruction.models.cascadenet.ccnn_block.CascadeNetBlock(model: Module, fft_centered: bool = True, fft_normalization: str = 'ortho', spatial_dims: Optional[Tuple[int, int]] = None, coil_dim: int = 1, no_dc: bool = False)[source]

Bases: Module

Model block for CascadeNet & Convolution Recurrent Neural Network.

This model applies a combination of soft data consistency with the input model as a regularizer. A series of these blocks can be stacked to form the full variational network.

forward(pred: Tensor, ref_kspace: Tensor, sens_maps: Tensor, mask: Tensor) → Tensor[source]

Forward pass of the model block.

Parameters

  • pred (Predicted k-space data.) – torch.Tensor, shape [batch_size, n_coils, height, width, 2]

  • ref_kspace (Reference k-space data.) – torch.Tensor, shape [batch_size, n_coils, height, width, 2]

  • sens_maps (Sensitivity maps.) – torch.Tensor, shape [batch_size, n_coils, height, width, 2]

  • mask (Mask to apply to the data.) – torch.Tensor, shape [batch_size, 1, height, width, 1]

Returns

torch.Tensor, shape [batch_size, height, width, 2]

Return type

Reconstructed image.

sens_expand(x: Tensor, sens_maps: Tensor) → Tensor[source]

Expand the sensitivity maps to the same size as the input.

Parameters

  • x (Input data.) – torch.Tensor, shape [batch_size, n_coils, height, width, 2]

  • sens_maps (Sensitivity maps.) – torch.Tensor, shape [batch_size, n_coils, height, width, 2]

Returns

torch.Tensor, shape [batch_size, n_coils, height, width, 2]

Return type

SENSE reconstruction expanded to the same size as the input.

sens_reduce(x: Tensor, sens_maps: Tensor) → Tensor[source]

Reduce the sensitivity maps to the same size as the input.

Parameters

  • x (Input data.) – torch.Tensor, shape [batch_size, n_coils, height, width, 2]

  • sens_maps (Sensitivity maps.) – torch.Tensor, shape [batch_size, n_coils, height, width, 2]

Returns

torch.Tensor, shape [batch_size, height, width, 2]

Return type

SENSE reconstruction.

training: bool

Module contents