NaN output in Measurement modules


I am relatively new and not yet very familiar with CellProfiler. At the moment, I try to identify nuclei and cell borders of cardiomyocytes and to measure intensities and area sizes afterwards. Identification and the gros of the measurements work fine, but I still have a little problem concerning a few measurements. Some of the output values are just NaN which, in my judgement, shouldn’t be.
I attached the pipeline I built and one of the cardiomyocyte images. I would be very glad if you had a quick look on it.


P.S.: One other question: Is it possible to change the outline colors in the EditObjectsManually module on my own? Currently, I work with heart cross sections stained in red (nuclei) and green (cell membranes). Sometimes, I find it quite difficult to distinguish if the object outlines are continuous or dashed. In other words, sometimes, it’s hard to determine if I already selected the objects or not. What I think of is just changing the outline color into white. I suppose it would be much more distinguishable. Is there any possibility to do that?

Auswertung_WGA_[3].cppipe (15.2 KB)

NRCM_Kernintensität_Zellgröße[3].cppipe (12.7 KB)

Hi David,

With regards to the NaN’s: The NaNs are for the cells you selected for deletion; that means that for the cells that are removed, there are no downstream measurements made. Also, in EditObjectsManually, you selected to retain the original numbering. Since the cells are numbered as before, a row is allocated for them and their non-existent measurements, and hence, NaN are used as placeholders in the data tables.

Not currently, but I think that’s a good idea; I’ll post it to our requested features list. In the meantime, you can press the “Hide image” button to toggle the background image, so you can see the outlines better.