Counting Goblet cells in bronchial respiratory epithelium in PAS Alcian Blue staining

Dear all,

I am trying to asses the number of goblet cells in respiratory epithelium using H&E slides. Alternatively the slides are stained with PAS or Alcian blue which stains the cytoplasm of the goblet cells red or blue.

Would anybody have any experience in doing this and would have a pipline to do this or would be willing to think along?.

Many thanks

JL

!!!Apologies, just realised that I need to upload a usable image, will do so tomorrow!!!

To give you an idea of what I’m looking at, the images below are non usable examples

PAS stained goblet cells (pink/red):
image
image nd

Alcian Blue stained goblet cells
image

This won’t work on your original images, but from your first image, it doesn’t seem too hard to separate the target cells, as far as I can tell.

setImageType(‘BRIGHTFIELD_H_E’);
setColorDeconvolutionStains(’{“Name” : “H&E default”, “Stain 1” : “Hematoxylin”, “Values 1” : "0.15903 0.96721 0.19804 ", “Stain 2” : “Eosin”, “Values 2” : "0.4989 0.84084 0.20996 ", “Background” : " 255 255 255 "}’);
selectAnnotations();
runPlugin(‘qupath.imagej.detect.nuclei.WatershedCellDetection’, ‘{“detectionImageBrightfield”: “Hematoxylin OD”, “backgroundRadius”: 0.0, “medianRadius”: 0.0, “sigma”: 5.0, “minArea”: 10.0, “maxArea”: 5000.0, “threshold”: 0.5, “maxBackground”: 2.0, “watershedPostProcess”: true, “cellExpansion”: 1.0, “includeNuclei”: true, “smoothBoundaries”: true, “makeMeasurements”: true}’);

I did not tweak the parameters to pick up some of the weaker or smaller cells, but I suspect it won’t be too bad with the full image.


For this example I used a small annotation to set the hematoxylin vector to “red” and the Eosin vector to the background purple.
Pete describes the process here for HDAB, but you can use it for any colors.

1 Like