Pigeons Trained To Spot Cancer


Stephen Luntz

Freelance Writer

clockNov 18 2015, 17:36 UTC
3781 Pigeons Trained To Spot Cancer
A pigeon, a touchscreen and a food pellet dispenser could be the future of diagnostic medicine. What would House say? Credit: Copyright Univ. Iowa/Wassermann Lab

It's possible for pigeons, with a little training, to distinguish breast tissue from tumors on biopsy slides. No hospital is known to have plans to replace highly qualified medical professionals with pigeons (Columba livia), although given the potential salary savings, it may only be a matter of time.

Professor Richard Levenson of the University of California, Davis, showed 16 pigeons touchscreen images of microscope slides of either benign or malignant breast tissue. Half the pigeons were rewarded with food for picking healthy slides, the other half for choosing slides showing cancer, while subgroups were given slides in monochrome or higher resolution. Levenson then tested their capacity to classify slides they had not seen before. The results are published in PLOS ONE


"The pigeons were able to generalize what they had learned, so that when we showed them a completely new set of normal and cancerous digitized slides, they correctly identified them," Levenson said in a statement. "Their accuracy, like that of humans, was modestly affected by the presence or absence of color in the images, as well as by degrees of image compression.”

"Pigeons' accuracy from day one of training at low magnification increased from 50 percent correct to nearly 85 percent correct at days 13 to 15,” Levenson said. The paper notes that this is "a task that can perplex inexperienced human observers who typically require considerable training to attain mastery.” Pigeons proved able to adapt their learning when confronted with images at unfamiliar magnifications.

Levenson then tested the pigeons' radiology skills and reported, “The pigeons also learned to correctly identify cancer-relevant microcalcifications on mammograms.” No comparisons of humans have been conducted under similar conditions, but Levenson told IFLScience that, "I gave one of the problem sets to our administative assistants, with verbal instructions as well as visual training, and after one hour of exposure they were certainly not as good as the birds after several days."


The pigeons were not perfect, though. Levenson admitted “they had a tougher time classifying suspicious masses on mammograms – a task that is extremely difficult, even for skilled human observers." Nevertheless, radiologists scored 80 percent on the "relatively subtle" slides used for this test, while the pigeons couldn't beat chance.

Professor Edward Wasserman, a co-author from the University of Iowa, added: "Research over the past 50 years has shown that pigeons can distinguish identities and emotional expressions on human faces, letters of the alphabet, misshapen pharmaceutical capsules, and even paintings by Monet vs. Picasso.” Intriguingly, the paper notes, “The pigeons proved to have an affinity for histopathology, as they achieved stable high performance as fast as or faster than with other visual discrimination problems studied in our Iowa laboratory.”

Most remarkably, if the assessments of multiple pigeons on each slide were added together – in a sort of pigeon democracy the authors call “flock-sourcing” – accuracy reached 99 percent.


The authors argue that while doctors will remain essential for diagnosis, “Our results suggest that pigeons can be used as suitable surrogates for human observers in certain medical image perception studies, thus avoiding the need to recruit, pay, and retain clinicians as subjects for relatively mundane tasks.” For example, testing against pigeons may be a way to speed up the development of computerized cancer diagnosis.

  • tag
  • cancer,

  • diagnosis,

  • pigeons,

  • malignant,

  • breast tissue