AI helps to detect up to 80% of disease in animals soon

Researchers at the University of Cambridge have developed a new technique for estimating the severity of pain in sheep, called the Sheep Pain Facial Expression Scale (SPFES). This is a pain assessment tool based on the expression on the face of the sheep.

The research team hopes that this advanced method will help improve the health of sheep as well as other animals.

"We have studied a number of people over the years," said Professor Peter Robinson, head of research . But earlier works on animal faces were made by Darwin. Darwin said that both humans and animals express emotions through similar behaviors. Therefore, we think that the movements on animal faces and people will have something in common. "

According to one expert, when the sheep are hurt, five things will appear on their faces: their cheeks are tight, the ears are folded forward, the lips are down, the nostrils change into a V shape and the eyes narrow.

Facial manifestations of sheep tell us whether they are painful.

A repository of 500 sheep photos was used to train the machine with initial tests showing that the machine can accurately estimate pain levels up to 80%.

"The most interesting part is that you can see the apparent similarity between the facial features of the sheep and the face when both of them are hurt ," said Dr Marwa Mahmoud, who was in the study.

However, 'normalizing' the face of sheep in this machine model is a difficult task. The other sheep's face in profile is completely looking at them and you can not tell the sheep to keep its face shut when taking pictures.

The final artificial intelligence system can be used to detect early animal illnesses such as rotting feet. This will allow for accelerated treatment and pain relief. By placing a camera on the gut and other areas where the sheep are gathering, the researchers hope that the system will be able to identify any sheep that are suffering from it.