Researchers develop AI for animal husbandry

Researchers from the Swiss Federal Institute of Technology in Zurich (ETH) and the University of Zurich have developed a method that uses Artificial Intelligence (AI) to help analyze the behavior of animals. The algorithm can simplify time-consuming behavioral studies in complex situations, further details of which can be found in a press release. Moreover, it could also help to monitor the wellbeing of animals in zoos, for example, as well as those selected to take part in animal studies.

Normally, researchers are forced to analyze large amounts of video material for animal behavior studies. Viewing and logging this data is a time-consuming process. In addition, movements and slight changes in the behavior of animals can escape the human eye. The research group headed up by Mehmet Fatih Yanik, Professor of Neurotechnology, have now automated the analysis of such recordings. The image analysis algorithm they have developed uses computer vision and machine learning processes, while it is also able to distinguish individual animals in complex animal communities and recognize even the smallest changes in behavior that may indicate, for example, curiosity, fear or harmonious interactions with other members of their species.

“By detecting adverse social interactions or the onset of disease early on, keepers can swiftly respond to improve conditions for the animals in their care”, the press release explains. In this context, the method is already being trialed in Zurich Zoo. The algorithm could also help in biological and medical studies to protect animal from excessive levels of stress, anxiety and discomfort.