Artificial Intelligence and Pain Recognition in Horses

With the rapid expansion of artificial intelligence (AI) in various sectors of our lives, it’s not surprising that AI is nudging its way into our barns. Livestream horse monitors and smart halters are already in use, and experts are now looking at using AI to help assess pain in horses.

Since the late 1800s, medical professionals have appreciated that facial expressions play an important role in assessing pain. In the past two decades, research in this field has exploded. We now know that facial expressions play a “crucial role as a primary nonverbal method for managing peer interactions and conveying information about emotional states,” according to Chiavaccini and colleagues.^*

The Mouse Grimace Scale paved the way for other pain scales in animals, including the Horse Grimace Scale (HGS).

“Studies using this scale have been published for horses with equine gastric ulcer syndrome as well as laminitis, showing that the HGS was a useful clinical tool for assessing and monitoring pain,” shared Kathleen Crandell, Ph.D., a Kentucky Equine Research nutritionist.

As pointed out in a recent review of pain recognition technologies in horses, the grimace scales were developed using a restricted number of “action units” retrieved from picture-based patterns.^* Research on grimace scales in horses, however, is limited because blinded studies cannot be ethically performed (e.g., comparing induced pain in horses that were and were not treated with pain medications). Further, grimace pain scales rely heavily on a subjective human component.

According to Chiavaccini and coworkers, a form of AI called automated pain recognition is an innovative technology that could prove useful for recognizing pain in horses based on image sensor and pain algorithms.

They said, “AI can radically change how we recognize and treat pain in nonverbal patients, including animals, with an immense impact on veterinary medicine and animal welfare.”

These systems can be further developed to create sophisticated models capable of analyzing not only facial expressions but also body postures, gestures, vocalizations, and physiologic parameters “to provide timely and effective pain management, pivotal in minimizing suffering and improving the overall quality of life for animals under our care.”

“Because horses can be stoic, several common conditions might be more painful than we now believe. Possible examples include gastric ulcers, colic, laminitis, and even osteoarthritis,” noted Crandell.

Osteoarthritis, a chronic, painful degeneration of joints, is the most common cause of lameness in horses. Current estimates suggest that 50% of horses older than 15 years and 80-90% of horses over 30 years suffer from osteoarthritis.^** Are we truly recognizing this pain, leaving horses untreated? This is where AI could help.

“There is currently no cure for osteoarthritis. Instead, joint discomfort is managed using an arsenal of therapies, which may include corticosteroids, orthobiologics, nonsteroidal anti-inflammatory drugs, and nutritional supplements.

“To ease painful, swollen joints try supplementing with hyaluronic acid (Synovate HA) or a combination of glucosamine, chondroitin and hyaluronic acid, such as KER-Flex, available in the United States,” recommended Crandell.

 ^*Chiavaccini, L, A, Gupta, and G. Chiavaccini. 2024. From facial expressions to algorithms: a narrative review of animal pain recognition technologies. Frontiers in Veterinary Science 11:1436795.

 ^**Arantes Baccarin, R.Y., S.R. Torquato Seidel, Y.M. Michelacci, et al. 2022. Osteoarthritis: a common disease that should be avoided in the athletic horse’s life. Animal Frontiers 12(3): 25–36.