How to count cells that are isolated from a group?

Sample image and/or code


Hello, I’ve got a lot of images of beta cell islets surrounded by isolated single cells. I’m trying to make a pixel classifier to quickly identify those single cells and count them, but I’m running into some trouble. I tried making a stock pixel classifier using qupaths pixel classifier tool (random tree), but it has a relatively high false positive and negative rate. I imagine something that can identify cells, then check the surrounding area for more cells would do the trick, but I’m not sure how to go about this.

If you’ve made it this far, I really appreciate your time. Let me know what you think.

Have a good day!

If you can detect them using Cell Detection or create objects in a similar way, you can use the Analyze-> Spatial analysis options.
Delaunay clusters will let you identify the size of clusters of similarly classified cells, so clusters of 1 would have no cells of the same type within X distance, where X is what you chose when you ran the Delaunay clusters.
This requires classifying the objects that you want to separate from the rest.

If you use a pixel classifier, you could size threshold during creation (to avoid things too large to be a single cell and too small to be a single cell), and then try something similar.

1 Like