cmp.stages.diffusion.reconstruction module
Reconstruction methods and workflows.
- class cmp.stages.diffusion.reconstruction.DipyReconConfig[source]
Bases:
traits.has_traits.HasTraits
Class used to store Dipy diffusion reconstruction sub-workflow configuration parameters.
- imaging_model
Diffusion imaging model (For instance ‘DTI’)
- Type
Str
- flip_table_axis
Axis to be flipped in the gradient table.
- Type
traits.List([‘x’, ‘y’, ‘z’])
- local_model_editor
List of reconstruction models
- Type
{False: ‘1:Tensor’, True: ‘2:Constrained Spherical Deconvolution’}
- local_model
Reconstruction model selected (See
local_model_editor
) (Default: True, meaning Tensor is performed)- Type
traits.Bool
- lmax_order
Choices of maximal order to use for Constrained Spherical Deconvolution
- Type
traits.Enum([2, 4, 6, 8, 10, 12, 14, 16])
- single_fib_thr
FA threshold
- Type
traits.Float(0.7, min=0, max=1)
- recon_mode
Can be “Probabilistic” or “Deterministic”
- Type
traits.Str
- mapmri
- Type
traits.Bool(False)
- tracking_processing_tool
- Type
traits.Enum(‘MRtrix’, ‘Dipy’)
- laplacian_regularization
Apply laplacian regularization in MAP-MRI if
True
(Default: True)- Type
traits.Bool
- laplacian_weighting
Laplacian regularization weight in MAP-MRI (Default: 0.05)
- Type
traits.Float
- positivity_constraint
Apply positivity constraint in MAP-MRI if
True
(Default: True)- Type
traits.Bool
- radial_order
MAP-MRI radial order (Default: 8)
- Type
traits.Int
- small_delta
Small data for gradient table (pulse duration) used by MAP-MRI (Default: 0.02)
- Type
traits.Float
- big_delta
Big data for gradient table (time interval) used by MAP-MRI (Default: 0.5)
- Type
traits.Float
- radial_order_values
Choices of radial order values used by SHORE
- Type
traits.List([2, 4, 6, 8, 10, 12])
- shore_radial_order
Even number that represents the order of the basis (Default: 6)
- Type
traits.Str
- shore_zeta
Scale factor in SHORE (Default: 700)
- Type
traits.Int
- shore_lambda_n
Radial regularisation constant in SHORE (Default: 1e-8)
- Type
traits.Float
- shore_lambda_l
Angular regularisation constant in SHORE (Default: 1e-8)
- Type
traits.Float
- shore_tau
Diffusion time used by SHORE. By default the value that makes q equal to the square root of the b-value (Default: 0.025330295910584444)
- Type
traits.Float
- shore_constrain_e0
Constrain SHORE optimization such that E(0) = 1 (Default: False)
- Type
traits.Bool
- shore_positive_constraint
Constrain the SHORE propagator to be positive (Default: False)
- Type
traits.Bool
- class cmp.stages.diffusion.reconstruction.MRtrixReconConfig[source]
Bases:
traits.has_traits.HasTraits
Class used to store Dipy diffusion reconstruction sub-workflow configuration parameters.
- flip_table_axis
Axis to be flipped in the gradient table.
- Type
traits.List([‘x’, ‘y’, ‘z’])
- local_model_editor
List of reconstruction models
- Type
{False: ‘1:Tensor’, True: ‘2:Constrained Spherical Deconvolution’}
- local_model
Reconstruction model selected (See
local_model_editor
) (Default: True, meaning Tensor is performed)- Type
traits.Bool
- lmax_order
Choices of maximal order to use for Constrained Spherical Deconvolution
- Type
traits.Enum([2, 4, 6, 8, 10, 12, 14, 16])
- single_fib_thr
FA threshold
- Type
traits.Float(0.7, min=0, max=1)
- recon_mode
Can be “Probabilistic” or “Deterministic”
- Type
traits.Str
- cmp.stages.diffusion.reconstruction.create_dipy_recon_flow(config)[source]
Create the reconstruction sub-workflow of the
DiffusionStage
using Dipy.- Parameters
config (DipyReconConfig) – Workflow configuration
- Returns
flow – Built reconstruction sub-workflow
- Return type
nipype.pipeline.engine.Workflow
- cmp.stages.diffusion.reconstruction.create_mrtrix_recon_flow(config)[source]
Create the reconstruction sub-workflow of the
DiffusionStage
using MRtrix3.- Parameters
config (DipyReconConfig) – Workflow configuration
- Returns
flow – Built reconstruction sub-workflow
- Return type
nipype.pipeline.engine.Workflow