As far as I know, there is no direct way or plugin can do this in Fiji. But of course you can achieve this using Fiji, with the assist of macros and scrips. Briefly, you have to use 'threshold' methods to binary green/red/neuron channels, then 'analysis particle' can be used to extract the signal's position information on each channel. Finally you have to write scripts (Javescript or Jython) to implement a loop function to check the overlay of green/red signals with neuron.

I also have some matlab scripts developed several years ago which can do similar stuff. Pls let me know if you are interested.

It could be slightly tricky depending on the type of signal. The first thing (and probably the difficult part) will be to segment cell bodies based on the green. Is the green fluorescence showing as speckles or a smear over the entire cell body? If the green is diffuse, you can use that to segment out cell bodies and then simply calculate how many of those have red in them. This will not be challenging in Fiji. Do you have a sample image that you can share/post here so that we get a better idea of the problem? Regards

Hi Tom, 

We had a user with exactly the same kind of need, perhaps you could provide me with an example dataset so I can see whether this method can be of use to you?

Best

Thank you for the help.

My solution:

I ended up selecting all the areas above a certain intensity using Image>Adjust>Threshold (note that this converts the channel to binary).  Then I did Analyze>Analyze Particles for the green channel and added the particles to the ROI Manager.  From there I was able to select the areas in the ROI manager and measure the Min/Max/Mean intensities in each of them in the red channel and a nonbinary duplicate of the green channel.  Later on I will do the analysis in MATLAB.

Another question:

I am still running into another problem.  I will soon be using channel rhodopsin bound EYFP and also staining for NeuN with a far red antibody.  This will result in many processes being labeled in addition to the cell bodies.  My PI wants me to figure out how to remove background and only measure intensity in the cell body by having the ROI for each cell in the green channel be defined by a circle around the NeuN labeled cell body that has a certain intensity (to be decided).  The antibody we use for the red channel mostly labels the cell bodies.  Do you know if there is a way to do this either in ImageJ/FIJI or in MATLAB?