Development and validation of an objective swine feet and leg conformation procedure using digital imagery
The objectives of this thesis were i) to develop and assess repeatability of an objective method for evaluating feet and leg conformation in high parity sows and ii) to validate and examine measurements using the developed objective method with a group of pedigreed gilts from age at selection through their first parity and into their second gestation. For the first objective, 45 multiparous sows (average parity 6.7 ± 2.5; parity range 5 to 14) from two commercial farms (n = 21 farm one and n = 24 farm two) were used. Sows were moved to a pen where digital images of the profile and rear stance were captured. On average, 2.8 and 8.1 final profile images were used per sow at farm one and farm two respectively. Farm two had over twice the number of profile images as farm one, as farm one was taken from the right side only whereas farm two had both left (average 4.2) and right (average 3.9) profile images. Additionally, 2.6 rear stance images were used for measurement per sow. A joint angle measuring system was devised to collect angle measurements on the knee, front and rear pastern, hock and rear stance. Joint measurements were analyzed using mixed model methods, including farm, side of measurement and parity as fixed effects and sow as a random effect. Intraclass correlation coefficients were calculated to evaluate process repeatability. No significant farm or parity differences were observed for joint angles measured except for the knee angle between farms (P < 0.05) and the hock angle between sows parity six and ≥ seven (P < 0.05). Side was significantly different in all joints measured (P < 0.05), except for in the rear stance measurement where side is not applicable. Joint angle measurement repeatability ranged from 0.58 to 0.87. Lowest and highest repeatabilities were observed for the knee and hock angle measurements, respectively. For the second objective, gilts were selected from a single population and moved to three different farms. Profile and rear stance images were obtained from gilts at selection (319; average age 21.6 ± 1.8 weeks; range 19 to 25) and during their second gestation (277; average gestation 26.7 ± 17.2 days; range 0 to 87). Knee, front and rear pastern, hock, and rear stance joint angles were measured using image analysis software. To evaluate symmetry and joint angle differences due to age between the same individual, only females with repeated measures at selection and post first parity, when second gestation days were between 0 to 21 (126 females), were used. Mixed model equations were used including parity (zero or one) and profile side (left or right) as fixed effects. Parity was included as a repeated variable with the animal as the subject. Knee and rear pastern angles decreased (weakened) and hock angles increased (straightened) as age progressed (P < 0.05). All joint measurements were symmetric between left and right legs (P > 0.05) except for the hock where a difference (P < 0.05) was observed. To evaluate gestation age effects on joint angles, only the measurements taken during the second gestation were used. Data was analyzed using mixed model equations including farm and side as fixed effects and gestation age as continuous covariate and animal was included as random effect. Farm was a significant source of variation for knee, front and rear pasterns, and rear stance angle measurements (P < 0.05). Additionally, asymmetry was detected in knee, and front and rear pasterns (P < 0.05). Front pastern and hock angles increased (straightened) as gestation age increased, while knee angle decreased (weakened) (P < 0.05). Heritability estimates were low to moderate for profile angles and was not estimable for the rear stance position. Results suggest that as age increases leg structure changes, with the rear leg joints showing greater variation from selection to first parity. Results also suggest that environmental factors such as farm where animals are housed could contribute to angle differences. Small angle changes in the front leg could indicate structure may change over the life of the animal; however, rear leg structure and its impact on longevity still require further investigation. Results from this body of work have set the ground work for an objective feet and leg joint conformation method using digital imagery. It is still necessary to look further into the life of the animal and understand the full genetic control over the change in structure until complete physical maturity and its association with lifetime productivity in the sow.