Question about small differences in results (same slide, same script)

Hello community
I am testing my script in the same TMA slide again, but now in the results I have small variations, but I have not changed any parameters of the script. Is this normal or what can it be due to?.

Thank you very much for your answers

(I’m working with M3 version)

I’ve just today learned this could happen when cell detection is applied to large regions that need to automatically be broken up into tiles; my guess is the overlaps aren’t resolved identically every time and so the cell counts could differ very slightly.

Might this fit with your case? Otherwise I’d really need to see the script or at least know what it does.


Dear Pete
What you say is correct that is the situation.
With the aim of evaluating reproducibility, we detect these small variations in positive detections with DAB using the same analysis script.
We compared the positive cell densities per mm2 with two different analysts who selected the ROIs according to the parameters we established. The correlation of cell densities is very good independent of the small variations of positive detections between the two analyzes performed (there are variations mainly in the number of cells by intensity 1+, 2+, 3+).

Best regards

1 Like

Thanks Carolina, that’s good to know.
I think the method of resolving overlaps still needs to be improved so that it is fully reproducible. But it’s good that the differences are small. I mostly work with smaller regions for which tiling isn’t necessary so hadn’t previously noticed any discrepancy.