Calculate/access RGB values for cytoplasm

hi guys,

I’m working with brightfield images and was wondering whether there is a way to calculate or access RGB values for the cytoplasm of detected cells.
I’m familiar with the option to do the calculate features of ROI ( select process cells) however I would need the cytoplasm only RGB values in the same manner.

I was thinking of a workaround calculating RGB values backwards via the 3 staining values, however I suppose there could be a more elegant (and efficient) way to that.

I appreciate your support.

Convert your RGB to a Composite image using Image>Color>Make Composite, add your ROI to the ROI Manager with ‘t’ and use the ROI Manager Multimeasue command. You’ll get one entry per r,g,b channel in the results table.
Jerome

If you only want the mean cytoplasmic values, you could subtract the whole cell and nuclear means based on their relative area (both of which can be generated with Add intensity features in the dropdown).

If you want per pixel values, you will probably need a mask, like what is used here to get the cytoplasm colocalization values.
https://gist.github.com/Svidro/68dd668af64ad91b2f76022015dd8a45#file-colocalization-of-channels-per-detection-0-2-0-groovy

That gets a bit complicated though, and as mentioned, I am not sure why the script fails on certain image types. Hopefully you simply need the cytoplasmic mean, despite how little that can often mean.

1 Like

that’s a creative and much more efficient workaround! thanks for sharing your ideas, I will probably try it this way as the mean values are what I’m currently looking for.
Thanks a lot.

(and also thanks to @jerome for your response, unfortunately I’m not familiar with imageJ and I’m just working with qupath at the moment.)

2 Likes