RunImageJ module treats differently pictures - Threshold problem


Hello everyone,
I am facing some problems with counting cells using CP and would like some help or a hint if somebody had the same issue.
What I am doing is staining tissue slides with DAPI and 3 markers and would like the software to count all possible combinations of positive cells. I have made a pipeline which I have attached but I realized there is a problem with the areas identified for the channels when calling RunImageJ module.
In details, I have a staining with DAPI, marker1, marker2, marker3 and I know from previous work that marker1 and marker2 are localized in the proximal part and not in the distal part of my tissue. Marker 3 is in both sites. I am taking pictures in confocal with the same settings and when acquiring the pictures, I clearly see that marker1 and marker2 are present in the proximal part and not in the distal. For some reason when I run the pipeline the software identifies an area for marker1 and marker2 while it should not and it gives me false positive nuclei for the distal picture. The pipeline works alright for proximal and intermediate parts of the tissue.
Part of the pipeline is to identify the staining for each channel and I am running the imagej to set the threshold. When the software does this step it adjusts the picture with a brightness/contrast value that brings up the background. I tried several things but it seems that they do not work. Does anybody know how I can fix that?
Somebody in a previous post said that CP sends pictures to ImageJ in an 8-bit format and pictures that are 16-bit are adjusted but I could not make it work. Is this the reason? Is there any way to fix this if this is the case?
Ideally, I would like to have the same pipeline for the different pictures because I could use different pipelines for all proximal and all distal pictures but I would not like to have that since I have many combinations of markers and then I will end up having several pipelines for each combination and that is not desired.
Thank you,
Counting PIPE.cpproj (106.4 KB)
Counting PIPE.cpproj (106.4 KB) (2.6 MB)


I’m afraid that sounds like something you need to address in whatever step(s) you’re running in ImageJ, not CP- CP just passes the images back and forth.

Can I ask why you’re using RunImageJ for this? CellProfiler does have an ApplyThreshold module that you could use to run your thresholding all in one program, and it’s got a lot of parameters you could try to tweak to set minimum and/or maximum values to use for thresholds, etc. Just a thought.