direct.nn.jointicnet package

direct.nn.jointicnet package#

Submodules#

direct.nn.jointicnet.config module#

class direct.nn.jointicnet.config.JointICNetConfig(model_name='???', engine_name=None, num_iter=10, use_norm_unet=False, image_unet_num_filters=8, image_unet_num_pool_layers=4, image_unet_dropout=0.0, kspace_unet_num_filters=8, kspace_unet_num_pool_layers=4, kspace_unet_dropout=0.0, sens_unet_num_filters=8, sens_unet_num_pool_layers=4, sens_unet_dropout=0.0)[source]#

Bases: ModelConfig

num_iter = 10#
use_norm_unet = False#
image_unet_num_filters = 8#
image_unet_num_pool_layers = 4#
image_unet_dropout = 0.0#
kspace_unet_num_filters = 8#
kspace_unet_num_pool_layers = 4#
kspace_unet_dropout = 0.0#
sens_unet_num_filters = 8#
sens_unet_num_pool_layers = 4#
sens_unet_dropout = 0.0#
__init__(model_name='???', engine_name=None, num_iter=10, use_norm_unet=False, image_unet_num_filters=8, image_unet_num_pool_layers=4, image_unet_dropout=0.0, kspace_unet_num_filters=8, kspace_unet_num_pool_layers=4, kspace_unet_dropout=0.0, sens_unet_num_filters=8, sens_unet_num_pool_layers=4, sens_unet_dropout=0.0)#

direct.nn.jointicnet.jointicnet module#

class direct.nn.jointicnet.jointicnet.JointICNet(forward_operator, backward_operator, num_iter=10, use_norm_unet=False, **kwargs)[source]#

Bases: Module

Joint Deep Model-Based MR Image and Coil Sensitivity Reconstruction Network (Joint-ICNet) implementation as presented in [1]_.

References: .. [1] Jun, Yohan, et al. “Joint Deep Model-Based MR Image and Coil Sensitivity Reconstruction Network (Joint-ICNet) for Fast MRI.” 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE, 2021, pp. 5266–75. DOI.org (Crossref), https://doi.org/10.1109/CVPR46437.2021.00523.

__init__(forward_operator, backward_operator, num_iter=10, use_norm_unet=False, **kwargs)[source]#

Inits JointICNet.

Parameters:

  • forward_operator (Callable) – Forward Transform.

  • backward_operator (Callable) – Backward Transform.

  • num_iter (int) – Number of unrolled iterations. Default: 10.

  • use_norm_unet (bool) – If True, a Normalized U-Net is used. Default: False.

  • **kwargs – Image, k-space and sensitivity-map U-Net models keyword-arguments.

forward(masked_kspace, sampling_mask, sensitivity_map)[source]#

Computes forward pass of JointICNet.

Parameters:

  • masked_kspace (Tensor) – torch.Tensor

  • shape (Sensitivity map of)

  • sampling_mask (Tensor) – torch.Tensor

  • shape

  • sensitivity_map (Tensor) – torch.Tensor

  • shape

Returns:

torch.Tensor Output image of shape (N, height, width, complex=2).

Return type:

out_image

direct.nn.jointicnet.jointicnet_engine module#

class direct.nn.jointicnet.jointicnet_engine.JointICNetEngine(cfg, model, device, forward_operator=None, backward_operator=None, mixed_precision=False, **models)[source]#

Bases: MRIModelEngine

Joint-ICNet Engine.

__init__(cfg, model, device, forward_operator=None, backward_operator=None, mixed_precision=False, **models)[source]#

Inits JointICNetEngine.

forward_function(data)[source]#

This method performs the model’s forward method given data which contains all tensor inputs.

Must be implemented by child classes.

Return type:

Tuple[Tensor, None]

Module contents#

Contents