American Scientist

To the Editors:

As a radiologist, I found Gene Tracy’s article “Learning To See” (July–August 2018) fascinating. In my residency I had to learn how to look at images of the human body and distinguish normal from abnormal, but it was much harder than that. There are all sorts of variations of “normal” that one has to learn before deciding that something is “abnormal.” Think of faces. No two faces are the same, but what constitutes an abnormal face? The same quandary confronts the radiologist.

After recognizing that there is, in fact, something abnormal present comes the second part of being a radiologist: making a diagnosis and describing what the abnormality means. The role of the diagnostic radiologist is to find a structural explanation for a patient’s symptoms. It’s sort of like playing “Where’s Waldo?” except often there is no abnormality present. A second and more insidious problem is satisfaction of search. After finding “Waldo,” diagnosticians must be careful not to miss other abnormalities that might be present.

Hank Gold
Lanesborough, MA