ants.segmentation.functional_lung_segmentation

functional_lung_segmentation(image, mask=None, number_of_iterations=2, number_of_atropos_iterations=5, mrf_parameters='[0.7,2x2x2]', number_of_clusters=6, cluster_centers=None, bias_correction='n4', verbose=True)[source]

Ventilation-based segmentation of hyperpolarized gas lung MRI.

Lung segmentation into classes based on ventilation as described in this paper:

Parameters:

  • image (ANTs image) – Input proton-weighted MRI.

  • mask (ANTs image) – Mask image designating the region to segment. 0/1 = background/foreground.

  • number_of_iterations (int) – Number of Atropos <–> bias correction iterations (outer loop).

  • number_of_atropos_iterations (int) – Number of Atropos iterations (inner loop). If number_of_atropos_iterations = 0, this is equivalent to K-means with no MRF priors.

  • mrf_parameters (str) – Parameters for MRF in Atropos.

  • number_of_clusters (int) – Number of tissue classes.

  • cluster_centers (numpy.ndarray or tuple) – Initialization centers for k-means.

  • bias_correction (str) – Apply “n3”, “n4”, or no bias correction (default = “n4”).

  • verbose (bool) – Print progress to the screen.

Returns:

  • Dictionary with segmentation image, probability images, and

  • processed image.

Example

>>> import ants
>>> image = ants.image_read(ants.get_data("mni")).resample_image((4,4,4))
>>> mask = image.get_mask()
>>> seg = ants.functional_lung_segmentation(image, mask, verbose=True,
                                            number_of_iterations=1,
                                            number_of_clusters=2,
                                            number_of_atropos_iterations=1)