Because modalities like MRI are non-ionizing and therefore not intrinsically harmful, I think it is reasonable to consider a wild extreme: in some future, what

Because modalities like MRI are non-ionizing and therefore not intrinsically harmful, I think it is reasonable to consider a wild extreme: in some future, what if a large group of people underwent imaging every month or every year. It's possible to imagine gaining a very good understanding of which lesions have malignant potential and which ones don't. The transition period that we are in now - where we are gaining information but not yet sure how to act on all of it - is painful. There are a lot of presumably unnecessary follow-up procedures. But it's possible that at some future point, we'll understand that 0.8mm nodules come and go throughout a lifetime and don't merit any action, whereas {some specific nodule in some specific location} almost always progresses to disease. Obviously what I'm describing is research, and so I'm not saying that we should treat clinical protocols differently right now. But I think it's not too hard to imagine that we can get to a point where we have a very good idea about which lesions to follow/treat and which lesions to leave be.