Individual pixel intensities within identified objects?

Hi there,

I’m trying to carry out a DNA content analysis using optical densitometry of stained nuclei. To find out DNA content I need to find the IOD for each nuclei.

Currently I have a pipeline that gives me a kind of surrogate of integrated optical density, by dividing the mean background intensity by the mean nuclear intensity, and taking a log10 of this number.

This isn’t the exact IOD however, which is the sum of (nuclear pixel intensity/mean background intensity) for each pixel in the nucleus.

I’m really struggling to find a way to do this and was wondering if anyone had any ideas? (I think I might need to find a way to get individual pixel intensity observations for each nuclear pixel?)

I’ve attached my pipeline and an example image.


LymphocyteOptimisedpipelineforbluefeulgen.cp (12.4 KB)

Hi Jacob,

For your purposes, I think you can use ImageMath. Use operation as Divide, the nuclear image as the 1st operand, and the mean background intensity for the 2nd operand, and uncheck the “Set values…” settings.

Since you are dividing a image matrix by a per-image value, the result will be an image matrix. You can then use MeasureObjectIntensity on the resultant image with the nuclei as the input object; the summation you want will be calculated as the per-object integrated intensity.


Hi Mark,

I managed to generate the measurement I needed through a series of imagemath modules, just like you’ve suggested. I now have a working pipeline!

Thanks for the reply, very much appreciated.