Issues with determining how many "live" and "dead" are on "DAPI"

cellprofiler
dapi
exporttodatabase
identifyprimary

#1

Hi, TEST.zip (17.4 MB)

So currently I am trying to figure out how to use this amazing software to its fullest, but I haven’t been able to,
My goal is to input pictures as three separate channels, and then identify out of all the ones that are DAPI, how many of them are green and how many of them are red, I have attached my pipeline and a sample image.
I always have to go through the pictures manually one by one because I havent been able to have a universal pipeline as the identify primary object seems to work differently for each picture (I am using manual but thats only because the imaging environment was the same and I wasnt able to utilize the other options)
I always end up with discrepancies in the final results, and I was hoping I would receive feedback on the modules I am using,
P.S. The images are of retinal photoreceptors

Sincerely,
Saud


#2

Hello Saud,
I would try introducing “RescaleIntensity” module for each channel in the beginning of the pipeline that should adjust intensity range of each image set independently and can help minimize the effect of its variability when thresholding. I would also avoid using the “Manual threshold” method. From your example images it looks like “RobustBackground” thresholding method should do well. See if these changes will improve the robustness of your pipeline.