Effect of video angle on detection of induced front limb lameness in horses

TL;DR

This study finds that the angle of a video recording affects how well horse lameness can be detected, with some angles providing more reliable assessments than others.

Contribution

The study identifies optimal video angles for detecting front limb lameness in horses using subjective and objective lameness assessments.

Findings

A video angle of the horse trotting directly towards the camera provides the best interobserver agreement for lameness assessment.

Objective lameness parameters strongly correlate with subjective scores (AUC of ROC = 0.87).

Some angles, like video angle 5, show perfect agreement for assessing sound horses.

Abstract

Lameness examinations are commonly performed in equine medicine. Advancements in digital technology have increased the use of video recordings for lameness assessment, however, standardization of ideal video angle is not available yielding videos of poor diagnostic quality. The objective of this study was to evaluate the effect of video angle on the subjective assessment of front limb lameness. A randomized, blinded, crossover study was performed. Six horses with and without mechanically induced forelimb solar pain were recorded using 9 video angles including horses trotting directly away and towards the video camera, horses trotting away and towards a video camera placed to the left and right side of midline, and horses trotting in a circle with the video camera placed on the inside and outside of the circle. Videos were randomized and assessed by three expert equine veterinarians using a 0–5 point scoring system. Objective lameness parameters were collected using a body-mounted inertial sensor system (Lameness Locator®, Equinosis LLC). Interobserver agreement for subjective lameness scores and ease of grading scores were determined.

Induction of lameness was successful in all horses. There was excellent agreement between objective lameness parameters and subjective lameness scores (AUC of the ROC = 0.87). For horses in the “lame” trials, interobserver agreement was moderate for video angle 2 when degree of lameness was considered and perfect for video angle 2 and 9 when lameness was considered as a binary outcome. All other angles had no to fair agreement. For horses in the “sound” trials, interobserver agreement was perfect for video angle 5. All other video angles had slight to moderate agreement.

When video assessment of forelimb lameness is required, a video of the horse trotting directly towards the video camera at a minimum is recommended. Other video angles may provide supportive information regarding lameness characteristics.

The online version contains supplementary material available at 10.1186/s12917-024-04032-9.

Figures