The quick answer, from extensive discussions I had over several years with many technical people, R&D and customers... although maybe not "authoritative": don't bother.
The longer answer:
Yes, it's possible to read the plate with a densitometer or plate reader if it can handle the lower contrast compared to a normal plate. The current gen of Kodak and Fuji plates have much better contrast than the first generations, making it fairly easy to identify the plate and separation. BUT - that's all it's really meant for. The contrast is created by a color-changing dye in the plate emulsion. That dye is NOT really part of the polymer that you print from though - it's just reacting to the same heat/IR light that cross-links the polymer, creating an image that's independent but readable. The answer I kept getting from these people was basically: "Yes, maybe you can read the dots... and maybe you can even read them accurately. BUT - they don't really mean anything in terms of calibration because the dye isn't linked to the printing of the final plate on press. Only by putting the plate on press and "developing" it will you get the exact dot size that will print. The edges of the dot may look dyed, but might wear off during the developing, rendering any measurement useless - especially in the highlights and shadows where you're having issues.
You can try to emulate the on-press development process yourself, but there's no guarantee that it's an accurate representation of the final on-press results. I've heard that Windex on a cloth will take off the unexposed emulsion, allowing for easier reading... but I'd suggest comparing the readings you get from that to actual on-press results to see if it's meaningful first.
Having said all that though... the whole point of these plates is that they largely make the need to read them regularly redundant. Being an ex-Kodaker and still strongly believing in their imaging technology, when combined with a high-res and consistent imaging head like the SquareSpot, once you calibrate it once (on press) it's a set-it-and-forget-it type technology. You're eliminating the major variable that people are calibrating for anyway - the processing steps and chemistry. ISO certifications and such may dictate that every plate be read, but those standards didn't envision a system that didn't have that large source of variation in the first place.
So - even if you CAN, it begs the question of SHOULD you. What is it you're looking to monitor, when you've eliminated the biggest problem already?
My $0.02... YMMV, and opinions abound I'm sure.
Kevin.