Israeli AI System Detects and Explains Pain in Horses to Improve Animal Care

Researchers at Tel-Hai College in Kiryat Shmona have developed an advanced artificial intelligence system that not only detects pain in horses but also identifies the specific anatomical areas causing discomfort. This breakthrough technology, detailed in the International Journal of Computer Vision, uses a novel method called SHIC-XE to provide consistent and anatomically accurate explanations of pain by mapping the AI's focus onto a fixed 3D model of the horse's face, overcoming challenges posed by movement and changing camera angles.

The AI was tested on datasets representing post-surgical pain, inflammatory-orthopedic pain, and acute mechanical pain, achieving high accuracy and statistically significant correlation with expert veterinarians’ assessments, particularly focusing on key anatomical regions such as the ears and cheek muscles. The research team, led by Dr. Marcelo Feigelstein, aims to bridge communication gaps between humans and animals by giving a "technological voice" to those unable to express pain verbally.

Beyond veterinary applications, the researchers see potential for this technology in human medicine, especially for patients unable to communicate their pain, such as premature infants, dementia patients, and sedated individuals. The geometric principle behind the AI’s ability to track anatomical regions despite movement could assist in clinical video analysis for pain assessment, neurological movement disorders, and surgical procedures.

The project involved collaboration between Tel-Hai College, the Tech4Animals lab, University of Haifa, and international partners from the Technion, University of Bern, University of Milan, University of São Paulo, and University of Newcastle. The team emphasizes that providing not only detection but also transparent explanations of AI decisions is crucial for building trust and making AI a reliable partner in medical and veterinary care.