Pigeons can identify breast cancer in medical images as accurately as human doctors.

A landmark 2015 study conducted by researchers at the University of California, Davis, and the University of Iowa revealed the surprising diagnostic capabilities of pigeons (Columba livia). The team, led by Professor Richard Levenson, trained 16 birds using operant conditioning to recognize microscopic signs of breast cancer. The pigeons were shown digitized slides of both benign and malignant human breast tissue and were rewarded with food for correct identifications.The results showed that after just two weeks of training, individual pigeons achieved an accuracy rate of 85% when identifying cancerous cells. However, when the researchers used a 'flock sourcing' approach—combining the independent responses of four different birds—the accuracy rate soared to an incredible 99%. This level of precision is comparable to that of professional radiologists and pathologists who spend decades in the field.Pigeons are uniquely suited for this task because their visual systems are remarkably similar to those of humans. They possess a high density of photoreceptors and can process complex visual patterns, including color, shape, and texture, with extreme speed. Beyond biopsy slides, the birds were also able to identify suspicious calcifications on mammograms, though they found classifying large masses more challenging.This research highlights the potential for using animal models to validate new medical imaging technologies before they are deployed in clinical settings. While pigeons will not be replacing doctors in hospitals anytime soon, their ability to detect subtle pathological features provides valuable insights into how visual processing works. The study remains a frequently cited example of how biological systems can rival sophisticated machine-learning algorithms in pattern recognition.