atlalign.non_ml package

Submodules

atlalign.non_ml.intensity module

Collection of intensity based registration methods.

antspy_registration(fixed_img, moving_img, registration_type='SyN', reg_iterations=(40, 20, 0), aff_metric='mattes', syn_metric='mattes', verbose=False, initial_transform=None, path=PosixPath('/home/docs/.atlalign'))[source]

Register images using ANTsPY.

Parameters

  • fixed_img (np.ndarray) – Fixed image.

  • moving_img (np.ndarray) – Moving image to register.

  • registration_type ({'Translation', 'Rigid', 'Similarity', 'QuickRigid', 'DenseRigid', 'BOLDRigid', 'Affine',) –

    ‘AffineFast’, ‘BOLDAffine’, ‘TRSAA’, ‘ElasticSyN’, ‘SyN’, ‘SyNRA’, ‘SyNOnly’, ‘SyNCC’, ‘SyNabp’,

    ’SyNBold’, ‘SyNBoldAff’, ‘SyNAggro’, ‘TVMSQ’, ‘TVMSQC’}, default ‘SyN’

    Optimization algorithm to use to register (more info: https://antspy.readthedocs.io/en/latest/registration. html?highlight=registration#ants.registration)

  • reg_iterations (tuple, default (40, 20, 0)) – Vector of iterations for SyN.

  • aff_metric ({'GC', 'mattes', 'meansquares'}, default 'mattes') – The metric for the affine part.

  • syn_metric ({'CC', 'mattes', 'meansquares', 'demons'}, default 'mattes') – The metric for the SyN part.

  • verbose (bool, default False) – If True, then the inner solver prints convergence related information in standard output.

  • path (str) – Path to a folder to where to save the .nii.gz file representing the composite transform.

  • initial_transform (list or None) – Transforms to prepend the before the registration.

Returns

  • df (DisplacementField) – Displacement field between the moving and the fixed image

  • meta (dict) – Contains relevant images and paths.

Module contents

Registration algorithm that do not use learning approaches.

Source

source/api/atlalign.non_ml.rst