Identifying labels which touch the background

In developmental biology it is a common task to differentiate layers of cells, for example in epithelial tissue. Therefore it might be useful to know if a cell is part of an outer layer, if it touches the background. In this notebook we measure and visualize this.

For demonstrating it we simulate a clumb of cells.

import pyclesperanto_prototype as cle

# import a function from a file in the same folder
from simulated_cell_clumb import simulate_data

cells = simulate_data()
cells

cle._ image shape (200, 200) dtype uint32 size 156.2 kB min 0.0 max 41.0

Determining which cells touch the background

To determine which cells touch the background, we need to produce a touch matrix which tells us which objects touch which others.

touch_matrix = cle.generate_touch_matrix(cells)
touch_matrix

cle._ image shape (42, 42) dtype float32 size 6.9 kB min 0.0 max 1.0

The first row and column in this image represent objects touching the background. We can read out this first row or column like this:

touching_background = touch_matrix[0]
touching_background

cle.array([0. 1. 1. 1. 0. 1. 1. 1. 0. 1. 0. 0. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 1. 1.
 0. 0. 1. 0. 0. 1. 0. 1. 1. 0. 0. 1. 1. 0. 0. 1. 1. 1.], dtype=float32)

And we can visualized it in the original image coordinates.

cle.replace_intensities(cells, touching_background)

cle._ image shape (200, 200) dtype float32 size 156.2 kB min 0.0 max 1.0

Or we can get a label image representing those objects.

cle.exclude_labels(cle.binary_not([touching_background]), cells)

cle._ image shape (200, 200) dtype uint32 size 156.2 kB min 0.0 max 20.0