Automated recognition of human facial expressions of pain and emotions is to a certain degree a solved problem, using approaches based on computer vision and machine learning. However, the application of such methods to horses has proven difficult. Major barriers are the lack of sufficiently large, annotated databases for horses and difficulties in obtaining correct classifications of pain because horses are non-verbal. This review describes our work to overcome these barriers, using two different approaches. One involves the use of a manual, but relatively objective, classification system for facial activity (Facial Action Coding System), where data are analyzed for pain expressions after coding using machine learning principles. We have devised tools that can aid manual labeling by identifying the faces and facial keypoints of horses. This approach provides promising results in the automated recognition of facial action units from images. The second approach, recurrent neural network end-to-end learning, requires less extraction of features and representations from the video but instead depends on large volumes of video data with ground truth. Our preliminary results suggest clearly that dynamics are important for pain recognition and show that combinations of recurrent neural networks can classify experimental pain in a small number of horses better than human raters.

Computer vision is a subcategory of artificial intelligence focused on extraction of information from images and video. It provides a compelling new means for objective orthopaedic gait assessment in horses using accessible hardware, such as a smartphone, for markerless motion analysis. This study aimed to explore the lameness assessment capacity of a smartphone single camera (SC) markerless computer vision application by comparing measurements of the vertical motion of the head and pelvis to an optical motion capture multi-camera (MC) system using skin attached reflective markers. Twenty-five horses were recorded with a smartphone (60 Hz) and a 13 camera MC-system (200 Hz) while trotting two times back and forth on a 30 m runway. The smartphone video was processed using artificial neural networks detecting the horse’s direction, action and motion of body segments. After filtering, the vertical displacement curves from the head and pelvis were synchronised between systems using cross-correlation. This rendered 655 and 404 matching stride segmented curves for the head and pelvis respectively. From the stride segmented vertical displacement signals, differences between the two minima (MinDiff) and the two maxima (MaxDiff) respectively per stride were compared between the systems. Trial mean difference between systems was $2.2$ mm (range 0.0–8.7 mm) for head and 2.2 mm (range 0.0–6.5 mm) for pelvis. Within-trial standard deviations ranged between 3.1–28.1 mm for MC and between 3.6–26.2 mm for SC. The ease of use and good agreement with MC indicate that the SC application is a promising tool for detecting clinically relevant levels of asymmetry in horses, enabling frequent and convenient gait monitoring over time.

It has been suggested that one of the underlying causes of asymmetrical performance and left/right bias in sound riding horses is laterality originating in the cerebral cortices described in many species. The aim of this paper is to review the published evidence for inherent biomechanical laterality in horses deemed to be clinically sound and relate these findings to descriptions of sidedness in equestrian texts. There are no established criteria to determine if a horse is left or right dominant but the preferred limb has been defined as the forelimb that is more frequently protracted during stance and when grazing. Findings on left-right differences in forelimb hoof shape and front hoof angles have been linked to asymmetric forelimb ground reaction forces. Asymmetries interpreted as motor laterality have been found among foals and unhandled youngsters, and the consistency or extent of asymmetries seems to increase with age. Expressions of laterality also vary with breed, sex, training and handling, stress, and body shape but there are no studies of the possible link between laterality and lameness. In a recent study of a group of seven dressage horses, a movement pattern in many ways similar to descriptions of sidedness in the equestrian literature, e.g. one hind limb being more protracted and placed more laterally than the other, has been documented. The role of innate laterality versus painful conditions, training, human handedness and simply habit remains to be determined. Understanding the biomechanical manifestations of laterality in healthy horses, including individual variation, would yield a potential basis for how laterality should be taken into account in relation to training/riding and rehabilitation of lameness.

Equine orthopedic pain scales are targeted towards horses with moderate to severe orthopedic pain. Improved assessment of pain behavior and pain-related facial expressions at rest may refine orthopedic pain detection for mild lameness grades. Therefore, this study explored pain-related behaviors and facial expressions and sought to identify frequently occurring combinations. Orthopedic pain was induced by intra-articular LPS in eight horses, and objective movement asymmetry analyses were performed before and after induction together with pain assessments at rest. Three observers independently assessed horses in their box stalls, using four equine pain scales simultaneously. Increase in movement asymmetry after induction was used as a proxy for pain. Behaviors and facial expressions commonly co-occurred and were strongly associated with movement asymmetry. Posture-related scale items were the strongest predictors of movement asymmetry. Display of facial expressions at rest varied between horses but, when present, were strongly associated with movement asymmetry. Reliability of facial expression items was lower than reliability of behavioral items. These findings suggest that five body behaviors (posture, head position, location in the box stall, focus, and interactive behavior) should be included in a scale for live assessment of mild orthopedic pain. We also recommend inclusion of facial expressions in pain assessment.

Background: Lameness evaluation of Standardbred trotters can be challenging due to discrepancies in observed movement asymmetry between in‐hand and track exercise, and between different trotting speeds. There are few studies on objective measurement of movement in Standardbreds, and little knowledge regarding biological variation and clinical significance of measured movement asymmetry in this breed.Objectives: To quantify the prevalence and magnitude of objectively measured movement asymmetry in young Standardbred trotters, and identify associations with trainer, sex, height, track type and in‐hand measurement prior to or after track trials.Study design: Cross‐sectional, observational study.Methods: A total of 114 Standardbred yearlings were evaluated with a wireless inertial sensor system during trot in‐hand and when driven on a track. After exclusions relating to lameness or technical difficulties, 103 horses were included in the study; 77 were evaluated in‐hand and on the track, 24 only in‐hand and 2 only on the track.Results: Front and/or hindlimb parameters were above asymmetry thresholds previously established for other breeds during in‐hand trials for 94 (93%) horses and during track trials for 74 (94%) horses. Most horses showed mild asymmetry. A minority of horses (20%) switched side of the asymmetry for one or more parameters between in‐hand and track trials. Mixed model analyses revealed no significant effects of trial mode (in‐hand or track trial, in‐hand trial pre‐ or post‐track trial, straight or oval track), trainer or horse height. Females had a significant but small reduction in asymmetry in one front limb parameter (HD_max) compared with males (1.7 mm, 95% CI 0.18‐3.28, P = .03).Main limitations: High data variability, reflected in large trial standard deviations, relating mainly to a lack of horse compliance.Conclusions: A high proportion of Standardbred yearlings showed movement asymmetries. There was no group‐level effect between in‐hand and track trials, however, considerable individual variation was observed.

Background: There is a paucity of research describing the gait pattern of lame horses at the walk. Objectives: To describe the changes in motion pattern and vertical ground reaction forces (GRFz) in horses with induced forelimb lameness at the walk and compare those changes with the changes observed at the trot. Study design: Experimental study. Methods: In 10 clinically sound Warmblood horses, moderate forelimb lameness was induced using a sole pressure model followed by trot and walk on a treadmill. Kinematic data were collected using 3D optical motion capture (OMC), and GRFz by an instrumented treadmill. Mixed models were used to compare sound baseline versus forelimb lameness (significance was set at P < .05). Results: Lameness induction significantly reduced peak GRFz on the second force peak, and vertical impulse in the lame limb. Stride and stance duration in all limbs were reduced. Lameness significantly affected the vertical movement symmetry of the head and withers. Maximum limb retraction angle, fetlock extension and protraction speed were reduced in the lame limb. Body centre of mass (COM) translation was reduced in the side‐to‐side direction and increased in the vertical and fore‐aft directions. Several compensatory kinetic and kinematic changes were observed in the nonlame limbs. The observed changes in both kinetics and kinematics were generally smaller at walk with fewer variables being affected, compared to the trot. Main limitations: Only one degree and type of orthopaedic pain (sole pressure) was studied. Conclusions: Compensatory strategies of forelimb lameness at the walk include alteration of several kinetic and kinematic parameters and have some specific patterns and inter‐individual differences that are not seen at the trot. However, much like at the trot, head movement and forelimb vertical force symmetry seem to be the most useful parameters to detect forelimb lameness at walk.