Probably, but if you have a description of what you actually want to do a or a screenshot of your existing rule, it would be much easier to help. It sounds like you could use the rule's Private Boolean (or any local--or I guess global--variable ... or any other method) to keep track of this and "reset" it when you're done. Something like this example from the Rule 4.0 documentation might be an example of what you're looking for:
I'm not sure how the rule in the previous post would help: the actions would still run (provided any conditions contained therein are met, as they likely would be above) any time either sensor changes to active or inactive, regardless of both sensors' previous states. You need something in the actions to say "hey, I've already run once" and a way to test for that and reset it when you're done (a purpose Private Boolean serves in this example). I may be missing or misunderstanding something, though.