The question is: If you have enough full body scans of many healthy people, and the statistical tools to model it (beyond "this range is OK"), whether this would reduce these false alarms to an acceptable level.

The real crux of it remains though: Let's say it finds something that increases your death risk by x=0.1%. Could you sleep? I'm not sure. Let's say the operation has 2x=0.2% risk. What do you do? What value of x makes this a problem for you?