Breeding herd retention poses significant challenges to the swine industry, with lameness ranking as a leading cause for breeding pig removal from the herd. Structural conformation is typically evaluated through visual observation; however, these methods are subjective and can vary in accuracy between observers. This study developed and assessed a computer-based algorithm for extracting feet and leg conformation traits from images of pigs. The algorithm, incorporating user intervention, identified the legs present in the images and obtained 21 characteristics from both the front and rear legs. The ability of each parameter to be inherited from parent to offspring was low to moderate across all traits. Genetic correlations between image traits and growth traits were generally low. However, male pigs with greater curvature, or a shorter distance between the ankle and shoulder, of the front leg tended to remain in the breeding herd for longer periods. This study highlights the feasibility of using image analysis to enhance selection for improved structural conformation in pig breeding, potentially addressing issues related to breeding herd retention and replacement rates.

Read the full article in the Journal of Animal Science.