A new study carried out by researchers at the University of California, Davis, and The University of Iowa, US, has revealed that pigeons can distinguish benign from malignant human breast histopathology and serve as promising surrogate observers of medical images.
The researchers conducted three experiments, in which the first experiment involved a cohort of eight pigeons, while second and third experiments included four pigeons each.
In the first experiment, researchers trained four pigeons with normal images and the other four with hue and brightness-balanced monochrome images.
In the second and third experiments, all of the training images have been of the same type for each of the four birds.
University of Iowa psychological and brain sciences professor and co-author of the study Edward Wasserman said: "The pigeons learned to discriminate benign from cancerous slides as fast in this research as in any other study we’ve conducted on pigeons in our laboratory.
"In fact, when we showed a cohort of four birds a set of uncompressed images, an approach known as ‘flock-sourcing’, the group’s accuracy level reached an amazing 99% correct, higher than that achieved by any of the four individual birds."
Housed in individual cages with ad-lib access to grit and water, all pigeons were maintained at 85% of their free-feeding weights by controlled daily rations.
For counterbalancing purposes, the pigeons in each quartet have been divided into two groups based on which subset of slides they were given during training.
In experiment one, a total of 144 images, including 48 at each of three magnification settings, taken from breast tissue samples were used.
UC Davis Health System pathology and laboratory medicine professor and lead author of the study Richard Levenson said: "With some training and selective food reinforcement, pigeons do just as well as humans in categorising digitised slides and mammograms of benign and malignant human breast tissue.
"The pigeons were able to generalise what they had learned, so that when we showed them a completely new set of normal and cancerous digitised slides, they correctly identified them.
"Their accuracy, like that of humans, was modestly affected by the presence or absence of colour in the images, as well as by degrees of image compression.
"The pigeons also learned to correctly identify cancer-relevant microcalcifications on mammograms, but they had a tougher time classifying suspicious masses on mammograms, a task that is extremely difficult, even for skilled human observers."
Each magnification grouping has been separated into two sets, including 12 benign and 12 malignant tissue examples (yielding sets A and B, respectively, for a total of 48 images).
Group one was trained with Set A and tested with Set B, while the opposite occurred for group two.
According to researchers, results from full-colour image training and testing established that pigeons could adeptly discriminate malignant from benign breast histological samples at multiple levels of magnification, in different spatial orientations, and effectively transfer those discriminations to novel stimuli.
Image: A pigeon being trained to screen images of benign and malignant breast tissue. Photo: courtesy of UC Davis Health System.