pywsi.segmentation package

Submodules

pywsi.segmentation.binary_segmentation module

pywsi.segmentation.binary_segmentation.gmm_thresholding(image_rgb)

Perform thresholding based on Gaussian mixture models.

The image is assumed to be a mixture of two gaussians. A lower mean sample belongs to the blobs while the higher mean shows the white background.

Parameters:

image_rgb: array_like

RGB input

Returns

——

gmm_threshold: float

GMM mean (minimum) of the two mixing populations

clf: sklearn.GaussianMixture

The entire sklearn model

pywsi.segmentation.binary_segmentation.poisson_deconvolve(h_channel_image)

Separate baclground and foreground intensities by doing a poisson partition.

For the i_th pixel h(i) = P0*P(i|0) + P1*P(i|1) where P0,P1 are prior probabilities of background and foreground.

For a threshold t, the parameters are given by: P(t) = sum_{i=0}^t h(i) mu_0(t) = 1/P0(t) * sum_{i=0}^t i*h(i)

P1(t) = sum_{i=t+1}^I_max h(i) mu_1(t) = 1/P1(t) * sum_{i=t+1}^I_max i*h(i)

Parameters:

h_channel_image: array_like

uint8 image (single channel)

pywsi.segmentation.diff_gaussian module

pywsi.segmentation.diff_gaussian.downsample(image, factor=2)

pywsi.segmentation.diff_gaussian.laplace_of_gaussian(input_image, foreground_mask, sigma_min, sigma_max, n_octave_levels=3)

pywsi.segmentation.fractal_dimension module

pywsi.segmentation.fractal_dimension.fractal_dimension(Z, threshold=0.9)

pywsi.segmentation.graph_cut module

pywsi.segmentation.graph_cut.perform_binary_cut(foreground, background)

Perform graph-cut based segmentation.

Parameters:

foreground, background: float32

Foreground, background probability matrix (w*h*3)

pywsi.segmentation.helpers module

pywsi.segmentation.helpers.collapse_labels(labelled_image)

Collapse labels in a labeleled image so that all labels are contigous

Parameters:

labelled_image: array_like

An image with labels

Returns:

label_collapsed_image: array_like

Image with contigous labels

pywsi.segmentation.helpers.collapse_small_area(labelled_image, minimum_area)

Collapse labelled image removing areas with too low are.

Parameters:

labelled_image: array_like

An image with labels

minimum_area: float

Areas with this and above area are retained

Returns

——-

label_collapsed_image: array_like

Image with contigous labels

pywsi.segmentation.max_clustering module

pywsi.segmentation.max_clustering.max_clustering(im_response, im_fgnd_mask, r=10)

Local max clustering pixel aggregation for nuclear segmentation. Takes as input a constrained log or other filtered nuclear image, a binary nuclear mask, and a clustering radius. For each pixel in the nuclear mask, the local max is identified. A hierarchy of local maxima is defined, and the root nodes used to define the label image. References [1] and [2]

Parameters:

im_response : array_like

A filtered-smoothed image where the maxima correspond to nuclear center. Typically obtained by constrained-LoG filtering on a hematoxylin intensity image obtained from ColorDeconvolution.

im_fgnd_mask : array_like

A binary mask of type boolean where nuclei pixels have value ‘True’, and non-nuclear pixels have value ‘False’.

r : float

A scalar defining the clustering radius. Default value = 10.

Returns:

im_label : array_like

im_label image where positive values correspond to foreground pixels that share mutual sinks.

seeds : array_like

An N x 2 array defining the (x,y) coordinates of nuclei seeds.

max_response : array_like

An N x 1 array containing the maximum response value corresponding to seeds.

References

[1]	XW. Wu et al “The local maximum clustering method and its application in microarray gene expression data analysis,” EURASIP J. Appl. Signal Processing, volume 2004, no.1, pp.53-63, 2004.

[2]	Y. Al-Kofahi et al “Improved Automatic Detection and Segmentation of Cell Nuclei in Histopathology Images” in IEEE Transactions on Biomedical Engineering,vol.57,no.4,pp.847-52, 2010.

Module contents

pywsi.segmentation.max_clustering(im_response, im_fgnd_mask, r=10)

Local max clustering pixel aggregation for nuclear segmentation. Takes as input a constrained log or other filtered nuclear image, a binary nuclear mask, and a clustering radius. For each pixel in the nuclear mask, the local max is identified. A hierarchy of local maxima is defined, and the root nodes used to define the label image. References [3] and [4]

Parameters:

im_response : array_like

A filtered-smoothed image where the maxima correspond to nuclear center. Typically obtained by constrained-LoG filtering on a hematoxylin intensity image obtained from ColorDeconvolution.

im_fgnd_mask : array_like

A binary mask of type boolean where nuclei pixels have value ‘True’, and non-nuclear pixels have value ‘False’.

r : float

A scalar defining the clustering radius. Default value = 10.

Returns:

im_label : array_like

im_label image where positive values correspond to foreground pixels that share mutual sinks.

seeds : array_like

An N x 2 array defining the (x,y) coordinates of nuclei seeds.

max_response : array_like

An N x 1 array containing the maximum response value corresponding to seeds.

References

[3]	XW. Wu et al “The local maximum clustering method and its application in microarray gene expression data analysis,” EURASIP J. Appl. Signal Processing, volume 2004, no.1, pp.53-63, 2004.

[4]	Y. Al-Kofahi et al “Improved Automatic Detection and Segmentation of Cell Nuclei in Histopathology Images” in IEEE Transactions on Biomedical Engineering,vol.57,no.4,pp.847-52, 2010.