intensity_normalization.util package

Submodules

intensity_normalization.util.coregister module

Co-register images with ANTsPy.

class intensity_normalization.util.coregister.Registrator(template: nib.nifti1.Nifti1Image | ants.ANTsImage = None, *, type_of_transform: str = 'Affine', interpolator: str = 'bSpline', metric: str = 'mattes', initial_rigid: bool = True)

Bases: SingleImageCLI

static description() → str

classmethod from_argparse_args(args: Namespace) → Registrator

static fullname() → str

classmethod get_parent_parser(desc: str, valid_modalities: frozenset[str] = frozenset({'flair', 'md', 'other', 'pd', 't1', 't2'}), **kwargs: Any) → ArgumentParser

static load_image(image_path: str | PathLike) → ants.ANTsImage

static name() → str

register_images(images: collections.abc.Sequence[nib.nifti1.Nifti1Image | ants.ANTsImage], /) → collections.abc.Sequence[nib.nifti1.Nifti1Image | ants.ANTsImage]

register_images_to_templates(images: collections.abc.Sequence[nib.nifti1.Nifti1Image | ants.ANTsImage], /, *, templates: collections.abc.Sequence[nib.nifti1.Nifti1Image | ants.ANTsImage]) → collections.abc.Sequence[nib.nifti1.Nifti1Image | ants.ANTsImage]

intensity_normalization.util.coregister.register(image: ValidImage, /, template: Optional[ValidImage] = None, *, type_of_transform: str = 'Affine', interpolator: str = 'bSpline', metric: str = 'mattes', initial_rigid: bool = True, template_mask: Optional[ValidImage] = None) → nib.nifti1.Nifti1Image | ants.ANTsImage

intensity_normalization.util.histogram_tools module

Process the histograms of MR (brain) images

intensity_normalization.util.histogram_tools.get_first_tissue_mode(image: ImageLike, /, *, remove_tail: bool = True, tail_percentage: float = 99.0) → float

Mode of the lowest-intensity tissue class

Parameters:

• image – array of image data (like an np.ndarray)

• remove_tail – remove tail from histogram

• tail_percentage – if remove_tail, use the histogram below this percentage

Returns:

mode of the lowest-intensity tissue class

Return type:

first_tissue_mode

intensity_normalization.util.histogram_tools.get_largest_tissue_mode(image: ImageLike, /) → float

Mode of the largest tissue class

Parameters:

image – array of image data (like an np.ndarray)

Returns:

value of the largest tissue mode

Return type:

largest_tissue_mode

intensity_normalization.util.histogram_tools.get_last_tissue_mode(image: ImageLike, /, *, remove_tail: bool = True, tail_percentage: float = 96.0) → float

Mode of the highest-intensity tissue class

Parameters:

• image – array of image data (like an np.ndarray)

• remove_tail – remove tail from histogram

• tail_percentage – if remove_tail, use the histogram below this percentage

Returns:

mode of the highest-intensity tissue class

Return type:

last_tissue_mode

intensity_normalization.util.histogram_tools.get_tissue_mode(image: ImageLike, /, *, modality: Modality) → float

Find the appropriate tissue mode given a modality

intensity_normalization.util.histogram_tools.smooth_histogram(image: ImageLike, /) → tuple[ImageLike, ImageLike]

Use kernel density estimate to get smooth histogram

Parameters:

image – array of image data (like an np.ndarray)

Returns:

domain of the pdf pdf: kernel density estimate of the pdf of data

Return type:

grid

intensity_normalization.util.io module

Input/output utilities for the project

intensity_normalization.util.io.gather_images(dirpath: str | PathLike, *, ext: str = 'nii*', exclude: Sequence[str] = ()) → list[pymedio.image.Image]

return all images of extension ext from a directory

intensity_normalization.util.io.gather_images_and_masks(image_dir: intnormt.PathLike, mask_dir: intnormt.PathLike | None = None, *, ext: str = 'nii*', exclude: collections.abc.Sequence[str] = ()) → tuple[PymedioImageList, PymedioMaskListOrNone]

intensity_normalization.util.io.glob_ext(dirpath: str | PathLike, *, ext: str = 'nii*', exclude: Sequence[str] = ()) → list[Path]

return a sorted list of ext files for a given directory path

intensity_normalization.util.io.split_filename(filepath: str | PathLike, /, *, resolve: bool = False) → SplitFilename

split a filepath into the directory, base, and extension .. rubric:: Examples

>>> split_filename("path/base.ext")
SplitFilename(path=PosixPath('path'), base='base', ext='.ext')

intensity_normalization.util.io.zip_with_nones(*args: Sequence[Any] | None) → Zipped

zip sequence args but if an arg is None, yield None in that argument index .. rubric:: Examples

>>> for x, y, z in zip_with_nones((1, 2), None, ("a", "b")):
...    print(x, y, z)
1 None a
2 None b

intensity_normalization.util.preprocess module

Preprocess MR images for image processing.

Preprocess MR images according to a simple scheme: 1) N4 bias field correction 2) resample to X mm x Y mm x Z mm 3) reorient images to specification

class intensity_normalization.util.preprocess.Preprocessor(*, resolution: tuple[float, ...] | None = None, orientation: str = 'RAI', n4_convergence_options: dict[str, Any] | None = None, interp_type: str = 'linear', second_n4_with_smoothed_mask: bool = True)

Bases: SingleImageCLI

static description() → str

classmethod from_argparse_args(args: Namespace) → Preprocessor

static fullname() → str

classmethod get_parent_parser(desc: str, valid_modalities: frozenset[str] = frozenset({'flair', 'md', 'other', 'pd', 't1', 't2'}), **kwargs: Any) → ArgumentParser

static load_image(image_path: str | PathLike) → ants.ANTsImage

static name() → str

intensity_normalization.util.preprocess.preprocess(image: intnormt.ImageLike, /, mask: intnormt.ImageLike | None = None, *, resolution: tuple[float, ...] | None = None, orientation: str = 'RAS', n4_convergence_options: dict[str, Any] | None = None, interp_type: str = 'linear', second_n4_with_smoothed_mask: bool = True) → tuple[mioi.Image, mioi.Image]

Preprocess an MR image

Preprocess an MR image according to a simple scheme: 1) N4 bias field correction 2) resample to X mm x Y mm x … 3) reorient images to RAI

Parameters:

• image – image to preprocess

• mask – mask covering the brain of image (none if already skull-stripped)

• resolution – resolution for resampling. None for no resampling.

• orientation – reorient the image according to this. See ANTsPy for details.

• n4_convergence_options – n4 processing options. See ANTsPy for details.

• interp_type – interpolation type for resampling choices: linear, nearest_neighbor, gaussian, windowed_sinc, bspline

• second_n4_with_smoothed_mask – do a second N4 with a smoothed mask often improves the bias field correction in the image

Returns:

preprocessed image and corresponding foreground mask

intensity_normalization.util.tissue_membership module

Find the tissue-membership of a T1-w brain image.

class intensity_normalization.util.tissue_membership.TissueMembershipFinder(hard_segmentation: bool = False)

Bases: SingleImageCLI

static description() → str

classmethod from_argparse_args(args: Namespace) → TissueMembershipFinder

static fullname() → str

classmethod get_parent_parser(desc: str, valid_modalities: frozenset[str] = frozenset({'flair', 'md', 'other', 'pd', 't1', 't2'}), **kwargs: Any) → ArgumentParser

static name() → str

intensity_normalization.util.tissue_membership.find_tissue_memberships(image: intnormt.ImageLike, /, mask: intnormt.ImageLike | None = None, *, hard_segmentation: bool = False, n_classes: int = 3) → mioi.Image

Tissue memberships for a T1-w brain image with fuzzy c-means

Parameters:

• image – image to find tissue masks for (must be T1-w)

• mask – mask covering the brain of image (none if already skull-stripped)

• hard_segmentation – pick the maximum membership as the true class in output

• n_classes – number of classes (usually three for CSF, GM, WM)

Returns:

membership values for each of three classes in the image

(or class determinations w/ hard_seg)

Return type:

tissue_mask

Module contents