recon_surf.N4_bias_correct¶

recon_surf.N4_bias_correct.get_brain_centroid(itk_mask)[source]¶

Get the brain centroid from a binary image.

Parameters:

itk_masksitk.Image: Binary image to compute the centroid of its labeled region.

Returns:

np.ndarray: Brain centroid.

recon_surf.N4_bias_correct.get_image_mean(image, mask=None)[source]¶

Get the mean of a sitk Image.

Parameters:

imagesitk.Image: Image to get mean of.
masksitk.Image, optional: Optional mask to apply first.

Returns:

meanfloat: The mean value of the image.

recon_surf.N4_bias_correct.get_tal_origin_voxel(tal, image)[source]¶

Get the origin of Talairach space in voxel coordinates.

Parameters:

talnpt.ArrayLike: Talairach transform matrix.
imagesitk.Image: Image.

Returns:

vox_originnp.ndarray: Voxel coordinate of Talairach origin.

recon_surf.N4_bias_correct.itk_n4_bfcorrection(itk_image, itk_mask=None, shrink=4, levels=4, numiter=50, thres=0.0)[source]¶

Perform the bias field correction.

Parameters:

itk_imagesitk.Image: N-dimensional image.
itk_maskOptional[sitk.Image]: Image mask. Defaults to None. Optional.
shrinkint: Shrink factors. Defaults to 4.
levelsint: Number of levels for maximum number of iterations. Defaults to 4.
numiterint: Maximum number if iterations. Defaults 50.
thresfloat: Convergence threshold. Defaults to 0.0.

Returns:

itk_bfcorr_image: Bias field corrected image.

recon_surf.N4_bias_correct.normalize_img(itk_image, itk_mask, source_intensity, target_intensity)[source]¶

Normalize image by source and target intensity values.

Parameters:

itk_imagesitk.Image: Input image to be normalized.
itk_masksitk.Image | None: Brain mask, voxels inside the mask are guaranteed to be > 0, None is optional.
source_intensitytuple[float, float]: Source intensity range.
target_intensitytuple[float, float]: Target intensity range.

Returns:

sitk.Image: Rescaled image.

recon_surf.N4_bias_correct.normalize_wm_aseg(itk_image, itk_mask, itk_aseg, target_wm=110.0, target_bg=3.0)[source]¶

Normalize WM image so the white matter has a mean intensity of target_wm and the background has intensity target_bg.

Parameters:

itk_imagesitk.Image: N-dimensional itk image.
itk_masksitk.Image | None: Image mask.
itk_asegsitk.Image: Aseg-like segmentation image to find WM.
radiusfloat, int, default=50: Radius of ball around centroid. Defaults to 50.
centroidnp.ndarray, optional: Image centroid. Defaults to None.
target_wmfloat | int: Target white matter intensity. Defaults to 110.
target_bgfloat | int: Target background intensity Defaults to 3 (1% of 255).

Returns:

normedsitk.Image: Normalized WM image.

recon_surf.N4_bias_correct.normalize_wm_mask_ball(itk_image, itk_mask=None, radius=50.0, centroid=None, target_wm=110.0, target_bg=3.0)[source]¶

Normalize WM image by Mask and optionally ball around talairach center.

Parameters:

itk_imagesitk.Image: N-dimensional itk image.
itk_masksitk.Image, optional: Image mask.
radiusfloat, int, default=50: Radius of ball around centroid. Defaults to 50.
centroidnp.ndarray: Brain centroid.
target_wmfloat | int: Target white matter intensity. Defaults to 110.
target_bgfloat | int: Target background intensity. Defaults to 3 (1% of 255).

Returns:

normalized_imagesitk.Image: Normalized WM image.

recon_surf.N4_bias_correct.options_parse()[source]¶

Command line option parser.

Returns:

options: Namespace object holding options.

recon_surf.N4_bias_correct.read_talairach_xfm(fname)[source]¶

Read Talairach transform.

Parameters:

fnamestr: Filename to Talairach transform.

Returns:

tal: Talairach transform matrix.

Raises:

ValueError: if the file is of an invalid format.