Reviewed by Lexie CornerNov 8 2024
Researchers aim to enhance pain recognition in children and other non-verbal patients by developing programs that can identify pain in animals, as discussed in a study published in Scientific Reports.
The subject of the study was a goat that presented with signs of distress, such as moping and grimacing, and exhibited reluctance to chew cud, indicating possible pain from a bladder stone—a common condition in small ruminants. While receiving treatment for the bladder stone, the goat contributed to ongoing research focused on accurately measuring pain in goats and other domestic animals, with the potential to extend these methods to humans in the future.
If we solve the problem with animals, we can also solve the problem for children and other non-verbal patients.
Ludovica Chiavaccini, DMV, DES., MS., Clinical Associate Professor, Anesthesiology, College of Veterinary Medicine, University of Florida
Chiavaccini and her colleagues captured images of both comfortable and distressed goats. They used this data to train an AI model capable of identifying goats in pain based solely on facial expressions.
The system, which has been trained and tested on 40 goats thus far, achieved an accuracy rate ranging from 62 % to 80 % in identifying faces associated with pain, depending on the testing methods used. With access to more data on goats and other animal species, these AI models could assist healthcare providers in effectively assessing and managing pain in patients who cannot communicate their discomfort.
Chiavaccini added, “It is not just an animal-welfare issue. We also know animals that are in pain don’t gain weight and are less productive. Farmers are becoming more and more aware of the need to control acute and chronic pain in animals.”
While further research is necessary, the implementation of AI-powered pain scales in veterinary clinics could address a longstanding challenge in animal care: accurately assessing animal pain. Historically, evaluating animal pain has been difficult and subjective. Veterinarians had to base their decisions on decades of experience. In an effort to lessen subjectivity, researchers have recently created standardized pain scales for various species; however, the quality of these measurements varies greatly.
At the time Chiavaccini and her team initiated their study, no specific pain scale for goats existed, despite a graduate student’s enthusiasm for the species. Currently, there is only one pain score for goats, but Chiavaccini pointed out that it has been validated solely for male goats undergoing castration, highlighting the need for a more comprehensive and widely applicable assessment system.
Journal Reference:
Chiavaccini, L. et. al. (2024) Automated acute pain prediction in domestic goats using deep learning-based models on video-recordings. Scientific Reports. doi.org/10.1038/s41598-024-78494-0