Cats can be extremely complex creatures at the best of times, and it can often be tricky to decipher their quirky behavior. This is especially true when it comes to pain. Due to instincts, cats tend to hide their pain well to not appear weak to competitors or potential predators. Translating this to modern domestic cats and telling whether a feline is in pain can be difficult, to say the least.
The rise of AI has seen exponential growth in recent years, and this extends to pets and their health, too. New AI-based apps promise to take some of the guesswork out of figuring out whether our feline friends are in pain.
The team at Sylvester.ai have developed an app called Tably, the world’s first AI app that aims to detect how a cat is feeling based on subtle facial cues and uses vet-approved pain scales to make an assessment. The tech is based on the Feline Grimace Scale, an effective tool used by scientists to determine pain in animals based on facial expressions. Using the app, owners simply take a photo of the cat’s face and can get a rating based on the AI models and integrated clinically validated veterinary pain scales.
Tably senior product manager Michelle Priest tells Wired, “With a high quality and full-face front image of the cat, the accuracy is 97% with adult cats of most breeds, which we are tremendously happy with.” The AI model is still learning to read kittens and some unusual breeds, although as more photos are added to the database, the AI should be able to learn to read these expressions better and better.
Similar to Tably, a team of researchers at the Tech4Animals lab from the University of Haifa is developing an AI-based app—Dr. Dolittl-E—to detect feelings of unease or illness in animals and hope to use AI tech to assist in veterinary healthcare. The team also used deep learning AI and facial recognition analysis to identify cats in pain, with a success rate of over 70%.
The research was based on a study of 29 female British Shorthair cats, and researchers took photos of the cat’s faces before and after sterilization, while they were still under the influence of painkillers, and again once the painkiller’s effects had subsided. (https://www.nature.com/articles/s41598-022-13348-1)
What Does This Mean for Cat Owners?
Can this technology actually be helpful to cat owners? Excitedcats.com in-house veterinarian Dr. Lorna Whittemore (BVMS) says it could potentially be: “Veterinary medicine is an ever-evolving field, and AI is likely to be increasingly helpful. Using human knowledge, compassion, and experience together with advanced technology will hopefully result in more cats being recognized as in need of pain relief and care.”
While this tech is certainly exciting and may help both cat owners and vets read cats better, there is still a long way to go. There are, obviously, some limitations, as while the app may tell owners if their cat is unhappy, they’ll still need a veterinarian to diagnose exactly what the issue is. There also may simply be nothing wrong; cats have bad moods too, just like humans, and may simply be hungry, tired, or grumpy.
Users of the app have had mixed experiences, with some cats perfectly happy and purring and the app giving back inaccurate results, as well as the opposite—a visibly irritated cat coming back with positive results.
Another excitedcats.com veterinarian, Dr. Luqman Javed (DVM), agrees. “Though probably useful, apps like these should be taken with a grain of salt. I’m not sure if they’d work if a cat had a facial deformity or injury. Plus, camera quality may matter, and even when in pain, cats’ faces might change if they see their owner approaching with a camera (especially if they associate it with treats). Also, some drugs meant to control pain can change face/eye appearance. While I do think such technologies look promising, they shouldn’t be used as a replacement for a veterinary consultation.”
While facial cues are certainly helpful, considering your cat’s entire body language, including, importantly, the tail, plus getting to know your cat’s moods, is arguably a more accurate representation of your cat’s feelings. Of course, the technology is still very early in development and may prove to be a useful tool for pet health.