Improving Positive cell count

Hello,
I’m running Positive Cell count and I’m not sure why is not identifying some cells like shown in my screenshot below. Is there any way I can improve it? I’m not interested in the cell count as I’m using the total area occupied by the fluorescence (thanks again for the script)

)

At a guess, I would think the objects it is missing are too large for your max area. Alternatively, Cell Detection is more of “Nuclear detection plus a cytoplasm expanded out from it” and won’t always deal well with objects that are not “nuclear shaped.”

If you want positive non-cytoplasmic staining, you may want to use a Thresholder instead.

https://qupath.readthedocs.io/en/latest/docs/tutorials/thresholding.html
You can run multiple thresholders inside one another, so one to detect tissue, one to detect positive staining inside of that, then overlap with another positive stain, etc. Or just create one for the green channel.

1 Like