Interacting with measurements in Napari#

When exploring data, it often makes sense to explore measurements as parametric images interactively.

import napari
from napari_simpleitk_image_processing import label_statistics, touching_objects_labeling
from napari_skimage_regionprops import add_table
from skimage.data import cells3d
from skimage.filters import gaussian
import pandas as pd

We start by loading a 3D image into napari and segmenting the nuclei.

import napari
viewer = napari.Viewer()

image = cells3d()[:,1]

viewer.add_image(image)
napari.utils.nbscreenshot(viewer)

blurred = gaussian(image, 1, preserve_range=True)
binary = blurred > 6000
labels = touching_objects_labeling(binary)

labels_layer = viewer.add_labels(labels)

napari.utils.nbscreenshot(viewer)

Feature extraction#

We next extract features using napari-simpleitk-image-processing and send the result as table to napari.

statistics = label_statistics(image, labels, shape=True)

pd.DataFrame(statistics)
label maximum mean median minimum sigma sum variance elongation feret_diameter flatness roundness equivalent_ellipsoid_diameter_0 equivalent_ellipsoid_diameter_1 equivalent_ellipsoid_diameter_2 equivalent_spherical_perimeter equivalent_spherical_radius number_of_pixels number_of_pixels_on_border
0 1 6497.0 6497.000000 6527.900391 6497.0 0.000000 6497.0 0.000000e+00 0.000000 0.000000 0.000000 0.0 0.000000 0.000000 0.000000 4.835976 0.620350 1 1
1 2 6070.0 6070.000000 6015.908203 6070.0 0.000000 6070.0 0.000000e+00 0.000000 0.000000 0.000000 0.0 0.000000 0.000000 0.000000 4.835976 0.620350 1 1
2 3 8583.0 6576.625000 6783.896484 4695.0 719.021339 315678.0 5.169917e+05 1.833872 11.874342 1.582815 0.0 2.712301 4.293070 7.872938 63.872162 2.254503 48 33
3 4 7160.0 6765.000000 7039.892578 6212.0 493.354842 20295.0 2.433990e+05 1.732051 1.414214 0.000000 0.0 0.000000 0.000000 0.000000 10.059235 0.894700 3 3
4 5 8962.0 6769.166667 6527.900391 4837.0 1310.103524 81230.0 1.716371e+06 2.515872 5.656854 1.573463 0.0 1.543809 2.429127 6.111373 25.347684 1.420248 12 12
... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...
457 458 8346.0 7196.000000 7039.892578 6591.0 791.127887 28784.0 6.258833e+05 1.000000 1.414214 0.000000 0.0 0.000000 0.000000 0.000000 12.185896 0.984745 4 4
458 459 8061.0 6887.750000 7295.888672 5548.0 1048.434508 27551.0 1.099215e+06 2.618034 2.236068 0.000000 0.0 0.000000 0.000000 0.000000 12.185896 0.984745 4 4
459 460 6734.0 6734.000000 6783.896484 6734.0 0.000000 6734.0 0.000000e+00 0.000000 0.000000 0.000000 0.0 0.000000 0.000000 0.000000 4.835976 0.620350 1 1
460 461 7350.0 6899.500000 7295.888672 6449.0 637.103210 13799.0 4.059005e+05 0.000000 1.000000 0.000000 0.0 0.000000 0.000000 0.000000 7.676633 0.781593 2 2
461 462 6165.0 6165.000000 6271.904297 6165.0 0.000000 6165.0 0.000000e+00 0.000000 0.000000 0.000000 0.0 0.000000 0.000000 0.000000 4.835976 0.620350 1 1

462 rows × 19 columns

labels_layer.features = statistics

add_table(labels_layer, viewer)

napari.utils.nbscreenshot(viewer)

Exercise#

Double-click a column header in the napari viewer and execute the next cell afterwards.

napari.utils.nbscreenshot(viewer)