A team of Israeli researchers has created an AI model capable of detecting pain in felines.
Big cats, such as lions, jaguars, cheetahs, leopards, and cougars, are known to hide their pain in order to avoid being targeted by other animals when they are weak and vulnerable. Interestingly, domestic cats exhibit the same behavior, which can make it difficult for even trained veterinarians to detect if they are suffering from pain. Sadly, many pet cats experience chronic pain that often goes unnoticed by their owners, leading to untreated conditions.
# Want to know if your cat is in pain?
Thankfully, researchers from the Tech4Animals lab at the University of Haifa, São Paulo University in Brazil, Lincoln University, and Nottingham University in the UK have developed innovative artificial intelligence (AI) that can identify pain in cats. By using deep-learning models and facial recognition analysis, the researchers achieved an impressive success rate of over 70% in identifying which cats were in pain. The study involved 29 short-haired British cats and the areas around the mouth and eyes were found to be the most significant in detecting pain.
This international study was recently published in the prestigious journal Scientific Reports from the Nature group, under the title “Automated Recognition of Pain in Cats.” It was led by doctoral student Marcelo Feighelstein and advised by Prof. Anna Zamansky and Prof. Ilan Shimshoni from the information systems department at the University of Haifa. This groundbreaking research offers new hope for the improved diagnosis and treatment of pain in pet cats, enhancing their overall well-being and quality of life.
# Developing the AI
 |
| Israeli Researchers Develop Method to Detect Pain in Cats. AI cat facial-pain examination technology (credit: University of Haifa research team) |
Facial expressions are recognized as one of the most distinct and reliable indicators of pain in both humans and animals. In fact, there has been a growing interest in developing tools for identifying pain-related facial expressions in various animal species since Dale Langford and colleagues at McGill University in Montreal first reported on the topic in mice. These tools have already been created and validated for rats, rabbits, horses, pigs, sheep, and ferrets, according to the researchers.
This breakthrough discovery involving cats has the potential to transform the way we care for our feline companions. With this AI model, anyone with a cat can easily take a photo and accurately check whether their pet is in pain without needing physical contact.
However, the researchers also acknowledged that using automated pain recognition in animals presents some challenges, which can also apply to recognizing emotions in general. The lack of data availability in the animal domain and the potential for greater facial variation in certain breeds can make population-level assessments problematic. Additionally, establishing a verbal basis for determining ground truth, which is commonly used in humans, is not applicable to animals.
Despite these challenges, the Tech4Animals Lab is developing a digital app called “Dr. Dolittl-E” that can detect animal emotions. This technology demonstrates how AI is advancing and transforming veterinary care, and it could help us gain a better understanding of our animal companions.