Dual Stain where both stains saturate to black

Any advice on how to approach these image analysis applications? In my image attached I have a green and red stain that both saturate to black. Depending on where I set my color vectors, I tend to have to compromise detection between these two colors.

Any tips or tricks on how to approach these images? Should I try to use a different color deconvolution algorithm to separate these colors? I’m open to FIJI or Qupath workflows. Or do I need to ask the wetlab to adjust the staining parameters to “lighten” things up?

At this point, all I care about is getting total area green and total area red.

Sorry I don’t have access to a higher quality image that I can distribute.

Any advice is appreciated!

If it is actually black (or close enough to 0,0,0), you have effectively lost the information. In some circumstances you might be able to determine based on the surroundings what that particular black “should” have been, but that requires controls and spatially separated colors. It does not look like you have that here. If either of the colors fluoresces strongly (I think fastRed does?), you might be able to overlay a fluorescent image and do some unmixing that way.

Or try to turn up the brightness/exposure on your light source until the darkest areas are more resolvable. Though the hematoxylin staining looks like it might be to faint for that if you are cell counting in QuPath.

I vaguely remember analyzing similar images before, and I am pretty sure I had to end up adding Intensity Measurements to the subcellular objects (with different color vectors), and then make assumptions about them based on those results. It was not a perfect world, though in my case I think almost all of the “black” was green.

1 Like

Thanks RA! You’ve gave me plenty of tips to work with.

Colour deconvolution does not work with neutral colours (greys, white, black).
Cranking up the brightness of the already captured image will convert black to grey… so you are stuck with the same problem again.
You can, however try if increasing a bit the illumination source allows to differentiate the saturated regions (so you need to capture the images again).
If that does not work, then there is too much chromogen/dye, and you should try staining for a shorter time, or at a lower temperature, or using lower concentrations of whatever is that you are using).