Cell counting and area measurement

I have images of GFP-expressing cells that are somewhat dim. However, the cells have a slightly brighter nucleus, and a bunch of processes extending from them. I started writing a pipeline, but it really wasn’t helping the image any. I managed to get primary objects (nuclei) identified, but I couldn’t for the life of me get a thresholding method that captured the processes without adding a whole bunch of garbage. The hope is to have total cell area per image and area/cells(nuclei) as outputs. Suggestions?


You probably have the best chance at cell identification if you have a separate nuclear stain to use as a basis. Is this an option in your case?


We’re trying to avoid it to allow for later assays as the cells are very precious. However, if we need to use a nuclear stain, we need to use a nuclear stain.