Efficient and sensitive animal pain detection approaches are increasingly studied with the goal of improving animal welfare and monitoring the efficacy of treatment and rehabilitation. The aim of this study was to determine the potential of various behaviours as sensitive indicators of subtle inflammation states in the equine animal. The long-term goal of this research is to understand how to objectively and remotely classify behaviours that are associated with inflammation using wearable inertial sensor technologies. This study represents a proof-of-concept investigation to ascertain what behavioural indices might be important in long-term monitoring of mild bilateral inflammation and recovery with a view to translating the approach to a technology-enabled remote monitoring paradigm. Bilateral synovitis of the intercarpal joints was induced in seven equines using lipopolysaccharide (0.25 ng) at time zero. Animals were confined to stables and monitored intermittently over seven days by video camera. White blood cell count, carpal circumference and food availability were recorded across the study. An ethogram was created to manually annotate behaviours from video footage following lameness induction at seven different timepoints across a 1-week period. Behaviour data were processed extracting the duration, frequency and variability of behaviours. One-way repeated measures ANOVA revealed a significant time effect for white blood cell count and behaviour switching. There were no significant changes in carpal circumferences and heart rate measures over the sampling period. Food availability appears to be an important contextual factor that should be considered in pain-related behavioural studies. We conclude that behaviour variability may be a promising indicator of subtle bilateral inflammation which should be further explored in a larger controlled trials and different pain presentations. Future work will seek to optimise grouping of behaviours associated with inflammation that can be detected using wearable technologies for future remote monitoring protocols.

越来越多地研究有效和灵敏的动物疼痛检测方法，以改善动物福利并监测治疗和康复的效果。这项研究的目的是确定各种行为作为马类动物细微炎症状态敏感指标的潜力。这项研究的长期目标是了解如何使用可穿戴惯性传感器技术对与炎症相关的行为进行客观和远程分类。这项研究代表了一项概念验证研究，旨在确定哪些行为指标在轻度双侧炎症和恢复的长期监测中可能很重要，以期将方法转化为技术支持的远程监测范例。在零时点用脂多糖（0.25 ng）在七个马中诱发腕间关节的双侧滑膜炎。将动物限制在马s中，并通过摄像机连续七天进行监测。在整个研究中记录白细胞计数，腕围和食物供应。创建了一个心电图，以便在1周的七个不同时间点对induction行诱导后的录像中的行为进行手动注释。处理行为数据以提取行为的持续时间，频率和变异性。单向重复测量方差分析显示白细胞计数和行为切换具有明显的时间效应。在采样期间，腕围和心率测量均无明显变化。可获得的食物似乎是与疼痛有关的行为研究中应考虑的重要背景因素。我们得出结论，行为变异性可能是微妙的双侧发炎的有前途的指标，应在较大的对照试验和不同的疼痛表现中进一步探讨。未来的工作将寻求优化与炎症相关的行为分组，这些行为可以使用可穿戴技术在将来的远程监控协议中进行检测。