Color threshold comparison among different sized images

Very new to ImageJ and after a lot of playing around I’m stuck - I have several hundred images which consist of a grayscale background with red-stained cells of interest and I would like to compare the amount of stained cells between images. I am thinking the best way to do this is set a color threshold to select just these cells and measure their area.

One issue I’m having is that my images are different sizes. Ideally I would like to select only colored cells within a particular bounded region of tissue (see attached images). Is there a way I can set these regions or crop the images so that they are standardized, ie the “area” of stained cells can be compared from one image relative to another?

Essentially I want to quantify “amount” of stained material for comparison between images, but for this to be meaningful it seems the region of interest has to be identical in size. Also, I’m assuming a macro would be helpful here, but I’m a bit lost in that regard.

If anyone has any suggestions of how to do this - or a more efficient way to go about it - please let me know, I’m very open to any input!

I think the best way is to split the RGB channels and work with the red one:
“Image >> Color >> Split Channels”

Perhaps even better is to look at the a or b channels after conversion to Lab:
“Image >> Type >> Lab Stack” and then “Image >> Stacks >> Stack to Images”

And yes, please learn how to use the mighty tool ImageJ by studying the User Guide.