Pigeons may distinguish between benign and malignant breast histology.

Pigeons may be able to distinguish between benign and malignant breast histology and radiology, according to an animal behavioral study published November 18, 2015 in the open-access journal PLOS ONE by Richard Levenson from the University of California Davis Medical Center, Edward Wasserman from the University of Iowa, and colleagues.

Share on Pinterest
The pigeons' training environment. The operant conditioning chamber was 150 equipped with a food pellet dispenser, and a touch-sensitive screen upon which the medical im- 151 age (center) and choice buttons (blue and yellow rectangles) were presented.
Credit: Levenson et al.

Pathologists and radiologists spend years acquiring and refining their medically essential visual skills. To better understand these skills, scientists trained several cohorts of pigeons, which share many visual system properties with humans, to view and identify benign and malignant histopathology and radiology images in a series of experiments.

After training with differential food reinforcement and controlling for various parameters, including image magnification, compression, and color and brightness, the birds quickly learned to distinguish benign from malignant human breast histopathology. Additionally, the pigeons were able to generalize what they had learned to novel image sets. The birds' histological accuracy, similar to humans, was modestly affected by the presence or absence of color, as well as by degrees of image compression, but the authors suggest these impacts could be ameliorated with further training. In radiology, the birds were similarly capable of detecting cancer-relevant microcalcifications on mammogram images. However, when given a different task--namely, classification of suspicious mammographic densities--the pigeons were only able to memorize the images and were not able to generalize to novel images.

The authors indicate that the birds' successes and difficulties suggest that pigeons may be well-suited to help scientists better understand human medical image perception, and may also prove useful in performance assessment and development of medical imaging hardware, image processing, and image analysis tools.