Seemed like a signal that there might be something there to be looked at - and there was: melanoma, caught very early and excised without complication. We never found out who the silent observer was, but we are grateful to him.
My guess (now updated with a short scan of recent papers) is that some melanoma presents in way that is fairly universally recognized. But that a meaningful fraction is "atypical" - meaning it is either not universally recognized (all that tremendous infighting among dermatologists hugely hampered getting large consistent databases needed to refine which characteristics are important), or it just plain doesn't have any visual characteristics that would differentiate it before it ulcerates or in some other way develops characteristics beyond purely visual differences.
PS Figured I'd look up the state of the state, and I think the above is a fairly good synopsis. The only thing to add is that the ABC method of diagnosis is (finally) falling out of favor since there appear to be several papers showing that people can't actually use it to find early melanomas.
Market Data 
Markets 
AI
