Ok, so I have a question for those of you that quantify your DAB stainings. I am quantifying my stains on a cells/mm2 basis, but I would also like to quantify the intensity of the stain. While doing a few test quantifications based off of previous posts on here, I have ran into a little bit of a conundrum.

Basically, I am using a positive control, a negative control, and an intermediate group to test how to quantify the intensity of the stains. Basically, if I split the channels/color calibrate to eliminate my hematoxylin stain, measure the mean grey value and calculate OD as OD=Log(255/mean grey value) I get very reasonable values.

Negative:

grey mean value (220)

OD (.064)

Positive:

grey mean value (174)

OD (.167)

What I do not quite understand is why the grey mean scale flips when I apply a threshold to the image. If I do a threshold and apply it to the both images, I get an Grey mean value of 1.6 (for negative) and 85 for positive. If I then calculate OD value I get 2.2 for negative and .47 for positive.

So what is the correct way to sort this out and why does thresholding flip my OD values? Would it be correct to report the this grey mean value after thresholding but not doing OD value, and if so, what would I call that?

Thanks,

Jordon

PS- I also understand that DAB doesn't exactly follow Beer-lambert for absorbance due, particularly for over developed stains. I'm merely looking to compare within a group of slides that have been stained on the exact same day.