A peer-reviewed study confirms the accuracy of Sleip’s motion analysis technology and its value in equine clinical assessments.

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 orthopedic gait assessment in horses using accessible hardware, such as a smartphone, for markerless motion analysis.

The study was led by Dr. Elin Hernlund and carried out by a group of researchers from the Swedish University of Agricultural Sciences, KTH Royal Institute of Technology and the engineering team at Sleip.

The study compared measurements of the vertical motion of head and pelvis obtained from a markerless system using Sleip, a smartphone single camera computer vision application, with those obtained using a multi-camera motion capture system with reflective markers attached to the horse's body. Simultaneous, synchronised recordings from both systems were compared.

The results showed that the mean difference between the two systems' measurement of lameness was below 2.2 mm, indicating that, apart from being easy to use, the smartphone tool can detect asymmetry in a horse’s gait at clinically relevant levels.

Twenty-five horses were recorded with a smartphone (60 Hz) and a multi-camera system (200 Hz) while trotting two times back and forth on a 30-meter runway. The smartphone video was then processed using artificial neural networks to detect the horse's direction, action, and motion of body segments. After filtering, the vertical displacement curves from the head and pelvis were synchronized between systems using cross-correlation. The study compared measurements of the vertical motion of head and pelvis obtained from a markerless system using a smartphone single camera computer vision application, with those obtained using a multi-camera motion capture system with reflective markers attached to the horse's body. Simultaneous, synchronised recordings from both systems were compared.

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.

‍