The result is the same with and without pointset

Hello ITKElastix team,

Thank you very much for your useful library. I want to try the 3D mask registration based on 4 points. My code is as follows. But when I try with and without a pointset, the output is the same.

for m in a:
    for n in b:
        fixed_image = itk.imread(f'mask_prepration_registration_corect/image{m}.nii.gz', itk.F)
        moving_image = itk.imread(f'mask_prepration_registration_corect/image{n}.nii.gz', itk.F)

        moving_mask = itk.imread(f'mask_prepration_registration_corect/mask{n}.nii.gz', itk.F)
        parameter_object = itk.ParameterObject.New()
        parameter_map_rigid = parameter_object.GetDefaultParameterMap('rigid')
        parameter_object.AddParameterMap(parameter_map_rigid)
        parameter_map_affine = parameter_object.GetDefaultParameterMap('affine')
        parameter_object.AddParameterMap(parameter_map_affine)
        parameter_map_bspline = parameter_object.GetDefaultParameterMap('bspline')
        parameter_object.AddParameterMap(parameter_map_bspline)

        result_image, result_transform_parameters = itk.elastix_registration_method(
            fixed_image, moving_image,
            fixed_point_set_file_name=f'data/landmarks/targets{m}.txt',
            moving_point_set_file_name=f'data/landmarks/targets{n}.txt',
            parameter_object=parameter_object,
            log_to_console=False)

        result_transform_parameters.SetParameter('FinalBSplineInterpolationOrder', '0')
        result_moving_mask = itk.transformix_filter(moving_mask, result_transform_parameters)

        itk.imwrite(result_moving_mask, f'masks_registration_result/{m}/reg_{m}_{n}.nii.gz')

Where is my solution wrong?

[sepidehamiri]

I believe you have to specify registration parameters that include point based metrics.

Developer docs on the metric: https://elastix.lumc.nl/doxygen/classelastix_1_1CorrespondingPointsEuclideanDistanceMetric.html

This link has some info and code from simpleelastix which is, N.B., not fully transferable but the principle is the same: https://simpleelastix.readthedocs.io/PointBasedRegistration.html

A good strategy may be to make to make two rigid parameter maps, the first with the corresponding points distance metric, and the second without. This workflow assumes your first rigid will use the corresponding points, find a good registration, then proceed with fully automatic approaches.

Heath

I added the following line and the problem was solved! Thank you very much for your help.

        parameter_map_rigid['Metric'] = [original_metric[0],
                                         'CorrespondingPointsEuclideanDistanceMetric']

[sepidehamiri]

Indeed, example 5 shows how to create a parameter map with the correct metrics in this case. Keep in mind that example 5 shows how to use a point set to help the registration of images, and example 9 shows how to transform the point sets.

You were absolutely right. I didn’t notice this difference.