Correction: Automated recognition of emotional states of horses from facial expressions.

Summary

# Editorial Summary: Automated Recognition of Emotional States in Horses from Facial Expressions Researchers at several institutions developed a machine learning system capable of identifying emotional states in horses by analysing facial expressions, addressing a significant gap in objective behavioural assessment tools available to equine professionals. Using computer vision and artificial intelligence algorithms trained on video footage of horses in various contexts, the team created an automated recognition system that could classify expressions associated with positive and negative emotional states. The technology demonstrated promising accuracy in distinguishing between distinct emotional phenotypes, offering farriers, veterinarians, and other practitioners a potentially objective method for assessing equine welfare beyond traditional visual observation alone. This automated approach could prove particularly valuable for monitoring pain responses post-treatment, evaluating stress during handling or training, and providing quantifiable data for research into equine behaviour and management practices. Whilst the correction note indicates methodological refinement was necessary, the underlying premise remains significant: developing reliable, technology-assisted tools for reading equine emotional responses could enhance clinical decision-making, improve animal welfare monitoring, and provide more rigorous evidence for evaluating management interventions across diverse equine settings.

Read the full abstract on PubMed

Key Findings

• •This is a correction notice to a previously published article on automated facial expression recognition in horses

Conditions Studied