Is my cat happy? This app predicts feline moods

“Is my cat happy or not?” Is a question that owners of notoriously obscure cats have been asking for ages – but which may soon be answered thanks to a new app.

The prototype app called Tablely (formerly Happy or Not) was published by Sylvester.AI – a joint venture between the Canadian artificial intelligence technology company Alta ML and The Bar G, a portfolio of companies in the fields of sleeping devices to ranching, animal health – and can tell how happy it is by looking at a photo of a cat by the expression on its face.

The app was launched after the developers of AltaML discovered the Feline Grimace Scale (FGS), a scientifically validated tool for assessing acute pain in cats based on facial expression changes published in Scientific Reports and Nature in 2019 with the idea of ​​using her Use machine learning expertise to promote healthier, happier lives for cats.

“The big challenge with cats is that they don’t express themselves when they are in pain,” said Chris O’Brien, vice president of product at AltaML. “You go and hide while a dog comes in and whines and sniffs you.”

Before the FGS, cats were considered delphic creatures that gave no clues when they were in pain. The truth is that these clues exist – they are just extremely subtle.

“Depending on the cat’s snout, eyes, ears, whiskers, or head position, you can tell whether a cat is in pain or not,” says Paulo Steagall, lead author of the FGS publication and associate professor of veterinary anesthesia and pain management. at the University of Montreal. The FGS was developed by Steagall and his team to be used by veterinarians, other veterinary technicians, and cat owners.

While the FGS is available for public use, Alta ML has coupled it with machine learning to make it more accessible to the general public. The app captures relevant points on a cat’s face, assigns them based on the FGS and outputs a measured value – happy or not – along with a percentage. The percentage does not correlate with the level of satisfaction, but with the level of confidence that the machine has.

“With a high-quality and full-surface front image of the cat, the accuracy is 97 percent per [with adults cats of most breeds], with whom we are very satisfied, ”says Michelle Priest, Senior Product Manager of the app.

According to Alta ML’s own survey of pilot app users, cat owners use it primarily to know how a cat is doing after an operation or to understand why a cat is hiding around the house all the time.

“It could be used to help decide whether to take the cat to the vet or even make an end-of-life decision,” says Priest. “You may have had a cat for 18 years, but you don’t want her to suffer. We also had this one guy who said he wanted to find out why his cat howled at night. “

Steagall made no comment on the validity of the app as he was not involved in creating it, but he says that anytime the FGS is used, context is everything. As an example, he takes a photo of his own cat taking a nap.

“She hears some background noise and has her eyes partially closed and her ears back,” he says. “So she got a high score on the grimace scale and said that you should probably give what is called emergency analgesia, but she was in no pain at all. She was just sleeping and paying attention to the background noise. “