Trained Pigeons Able to Identify Breast Cancer Tissues, Dogs Trained to Identify Prostate Cancer

A remarkable finding based on three experiments indicate that pigeons can have the ability to look at x-rays and microscope slides and determine which in the images are breast cancer tissue and which are the healthy ones. Not only are they able to tell, but they are able to make the distinction with 99 percent accuracy.

Pigeons, which are characterised by an exceptional ability to survive even the busiest of cities, depended on for many years to ferry messages and important documents across long distances and now even popular racers, seem to then have many other untapped potentials.

The study, which utilised a series of three tests, saw the test pigeons progress in learning the task of distinguishing between breast cancer tissues and images of healthy breast tissues.

The research, which was led by Professor Richard Levenson of the University of California Davis Medical Center, found that pigeons and humans share a number of visual system properties.  

The first test had eight pigeons view 144 images of breast tissue, which had different magnifications and were either coloured or uncoloured. Food reward and reinforcement were used with the pigeons whenever the pigeons identified the images correctly.  Incorrect identification meant that the pigeons had to view the images again and again until they were able to classify the images.  The pigeons used the yellow and blue buttons that flanked the images to give their answer.

Professor Levenson relayed, "With some training and selective food reinforcement, pigeons do just as well as humans in categorizing digitized slides and mammograms of benign and malignant human breast tissue."

Over a training period of 15 days, the birds' learning grew from 50 percent to 85 percent. At this point the pigeons were given new breast cancer tissue and healthy tissue images for assessment to eliminate memorisation as a possible cause for identification and accuracy.  The pigeons' learning was affirmed when they made correct identifications among familiar images with 87 percent accuracy and among new images with 85 percent accuracy. Using flock sourcing as basis, the birds were found to have a combined accuracy rate of 99 percent.

"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."

In previous studies, a similar training for dogs enabled the dogs to identify prostate cancer and did so with 90% accuracy.

See Also:

Real Time Analytics