Cambridge University scientists develop new AI that can tell if a sheep is in physical pain

London: The researchers at the University of Cambridge in the UK have devised a new artificial intelligence (AI) computer system that can determine whether a sheep is happy or sad by analysing its expressions.


They used five different facial expressions to estimate the severity of pain in sheep. In a large herd, the system could pick out sheep that are in distress. It also picks up common diseases such as foot rot and mastitis, a painful udder infection. This pain assessment would also help with early diagnosis.


The device could be used on other types of animals, such as rodents used in animal research, rabbits or horses.


About 500 images of sheep were used to train the AI system and early tests show the model is able to assess pain levels with 80% accuracy.


The different parts of the sheep’s faces have been examined on each photograph and ranked the pain levels on a scale of one to 10 of the animals according to SPFES the Sheep Pain Facial Expression Scale.



Researchers said while the results with still photographs have been successful, in order to make the system more robust, they require much larger datasets.


Professor Peter Robinson, of Cambridge’s Computer Laboratory, said that. ‘‘If they are able to train the system well enough, a camera could be positioned at a water trough or other place where sheep congregate, and the system would be able to recognise any sheep which were in pain.”


He added.”The farmer would then be able to retrieve the affected sheep from the field and get it the necessary medical attention.”