Can pigeons replace radiologists?

At the annual meeting last week of the Radiological Society of North America in Chicago, one expert in imaging was conspicuous by its absence - the pigeon.

Known for their loyalty, attention to detail, and sense of direction, and trusted to deliver the message, pigeons and radiologists have much in common. A study published in PLOS One showed pigeons can be trained to find cancers in mammograms. Researchers showed pigeons abnormal mammograms and rewarded them for spotting cancer. With time, pigeons became better at seeing cancer. As important, pigeons improved without doing a residency in radiology or even attending medical school.

Pigeons have been off our radar but are nifty creatures. During World War II, behavioral scientist B.F. Skinner proposed using pigeons to guide aerial bombing. Pigeons would peck incessantly when the target to be bombed moved away from the center of the screen, and kept pecking until the target returned to the center. The project was abandoned because no one took pigeons seriously.

For radiologists, who are used to hearing that computers can do our jobs, or that our work can be outsourced to India, pigeons may seem like a parody. But animals, at least those that can replace us, say a lot about us and how we practice our trade.

Radiologists are doctors who interpret medical imaging. The interpretation needs a medical degree because there is more to medical imaging than the images. It is not odd for an expert in imaging to concede that - there is more to flying a plane than the control panel.

Images can be useless without knowing why the patient is having the test. That is because diagnoses are not black and white on imaging, as there are shades of gray. These gray shades become brighter when we know more about the patient.

Doctors have an advantage over pigeons when it comes to patients. Pigeons are not known for their bedside manner. Even a positive finding on imaging must be seen in the context of the patient.

For example, if an MRI finds a small mass in the kidney in an 80-year-old woman who has a life expectancy of three years, the doctor can reassure her that even if the mass is a cancer, it is likely to be an overdiagnosis - which is a cancer that won't affect her survival. The doctor can counsel her that she should not worry about the small mass, but should enjoy whatever time she has left on this beautiful planet with her grandchildren. Such nuance may escape a pigeon.

Paradoxically, the real value of doctors', rather than pigeons', reading your CAT scans is not that doctors see every little abnormality, but that doctors know which findings may be ignored because they are likely not to harm you.

The trouble is that today, we practice medicine and diagnostic imaging to catch every single possibility in everyone, and we find small abnormalities, such as tiny masses in the thyroid gland, which will never harm the patient. But we cause a lot of grief to patients by finding small abnormalities.

The real threat to doctors are not pigeons, but thinking like pigeons. Using our medical degrees means ignoring abnormalities on CAT scans that may not harm the patient, as well as finding the abnormalities that will harm the patient.

The difference between radiologists and pigeons is that radiologists know when bad news may not necessarily be such bad news.

Saurabh Jha, MBBS, is an assistant professor of radiology at the Hospital of the University of Pennsylvania who has published regularly on the problem of overdiagnosis.

Contact him on Twitter @RogueRad.