Scientists Test AI That Can Identify Pain In Goats From Their Facial Expressions

Seeing pain in animals is not quite as easy as wondering how they feel; they can’t talk after all. There are ways to identify signs of pain, however—animals may express pain with unpleasant yelps, or not eat as expected. Judging that they are in pain isn’t always easy though, and it can be subjective, requiring decades of experience to make a judgment call about whether an animal is in distress.

Thankfully, for all its faults, one area where artificial intelligence excels is in pattern recognition. ChatGPT is actually pretty good at figuring out how to generate sentences that make sense after seeing lots of examples of them; Feed Google Lens a picture of a cat and it can find millions more, automatically sorting out all the cat’s unique features.

Scientists believe that the same concept can be applied to the perception of pain on the faces of animals. Decades of experience from veterinarians identifying painful faces can be fed into AI algorithms to automate the process for farmers. The technology may be used for some non-verbal patients, especially young children.

Phys.org reported this week on work from the University of Florida’s College of Veterinary Medicine. The researchers there focused on goats, photographing the faces of 40 subjects, some in pain and others in peace. They then fed the data to an artificial intelligence model that learned to distinguish pain based on facial expressions alone.

The algorithm they created is said to be 62% to 80% accurate in identifying facial pain. More data will be needed to improve accuracy, but what doctors say is very important if they can easily detect pain in non-verbal patients without relying on gut feeling.

“It’s not just an animal welfare issue,” said Ludovica Chiavaccini, clinical associate professor of anesthesiology at the University of Florida’s College of Veterinary Medicine. “We also know that animals in pain do not grow and produce well.” You can imagine that farmers have a program on a mobile device that allows them to quickly check many goats to see which ones need more testing without having to stop and slowly check each animal by hand or wait until they see serious problems.

It’s one of those areas where AI has an obvious role to play in automating something that usually takes humans a lot of time and experience to master. Computers are very adept at importing and learning vast amounts of information in a way that humans cannot easily do.