Deep learning pose detection model for sow locomotion.

Lameness affects animal mobility, causing pain and discomfort. Lameness in early stages often goes undetected due to a lack of observation, precision, and reliability. Automated and non-invasive systems offer precision and detection ease and may improve animal welfare. This study was conducted to create a repository of images and videos of sows with different locomotion scores. Our goal is to develop a computer vision model for automatically identifying specific points on the sow's body. The automatic identification and ability to track specific body areas, will allow us to conduct kinematic studies with the aim of facilitating the detection of lameness using deep learning. The video database was collected on a pig farm with a scenario built to allow filming of sows in locomotion with different lameness scores. Two stereo cameras were used to record 2D videos images. Thirteen locomotion experts assessed the videos using the Locomotion Score System developed by Zinpro Corporation. From this annotated repository, computational models were trained and tested using the open-source deep learning-based animal pose tracking framework SLEAP (Social LEAP Estimates Animal Poses). The top-performing models were constructed using the LEAP architecture to accurately track 6 (lateral view) and 10 (dorsal view) skeleton keypoints. The architecture achieved average precisions values of 0.90 and 0.72, average distances of 6.83 and 11.37 in pixel, and similarities of 0.94 and 0.86 for the lateral and dorsal views, respectively. These computational models are proposed as a Precision Livestock Farming tool and method for identifying and estimating postures in pigs automatically and objectively. The 2D video image repository with different pig locomotion scores can be used as a tool for teaching and research. Based on our skeleton keypoint classification results, an automatic system could be developed. This could contribute to the objective assessment of locomotion scores in sows, improving their welfare.

Trade-offs in the externalities of pig production are not inevitable.

Farming externalities are believed to co-vary negatively, yet trade-offs have rarely been quantified systematically. Here we present data from UK and Brazilian pig production systems representative of most commercial systems across the world ranging from 'intensive' indoor systems through to extensive free range, Organic and woodland systems to explore co-variation among four major externality costs. We found that no specific farming type was consistently associated with good performance across all domains. Generally, systems with low land use have low greenhouse gas emissions but high antimicrobial use and poor animal welfare, and vice versa. Some individual systems performed well in all domains but were not exclusive to any particular type of farming system. Our findings suggest that trade-offs may be avoidable if mitigation focuses on lowering impacts within system types rather than simply changing types of farming.

Lameness in Pregnant Sows Alters Placental Stress Response.

Pregnant sows from commercial pig farms may experience painful states, such as lameness, an essential indicator in assessing sow welfare. We investigated the effect of lameness during the last third of pregnancy on reproductive performance and placental glucocorticoid concentrations in sows. Periodic locomotion assessments were carried out on two commercial pig farms using a validated 0-5 scoring system (from 0: normal locomotion; to 5: a downer animal). Sows from both farms (N = 511) were grouped based on their average locomotion scores. On Farm 1, 30 sows were selected and grouped as either Not Lame (NL = 16; X¯ = 0-1) or Lame (L = 14; X¯ > 1). On Farm 2, 39 sows were selected and grouped as either Not Lame (G1 = 12; X¯ = 0-1), Moderately Lame (G2 = 13; X¯ = 1.1-2), or Severely Lame (G3 = 14; X¯ ≥ 2.1). Reproductive data (gestation length, litter weight, average piglet weight, litter size, and the number of piglets born alive/mummified/stillborn) were recorded on both farms. Moreover, on Farm 2, piglet intrauterine growth restriction score and the number of piglets dead during the first week were also recorded, and placenta samples were collected to determine their cortisol/cortisone concentrations. A linear mixed model was used to analyze the data. The proportion of lameness in pregnant sows (N = 511) was >40%, and the gestation length tended to decrease with the presence of lameness (p < 0.1) in both farms. G2 sows had a higher placental cortisol/cortisone ratio than G1 and G3 sows (p < 0.01). In conclusion, lameness was high in the sows assessed, which may decrease sow gestation length and reduce placental efficiency in protecting the offspring from the sows' stress response.

Asinine milk mitigates stress-mediated immune, cortisol and behavioral responses of piglets to weaning: A study to foster future interventions in humans.

Introduction: The present study assessed whether asinine milk supplementation improved the immune and behavioral responses of piglets during an early life weaning stress event as a model for its future use in humans. Methods: For this, 48 piglets from 4 different litters were used. At 20 days of age, piglets were weighed and allocated with their litter and dam into group pens until 28 days of age. Four piglets from each litter were then randomly assigned to either (1) asinine milk supplementation (n = 16) (2), skimmed cow milk supplementation (n = 16) or (3) no supplementation (n = 16; control group). The supplementations were voluntarily administered for 3 days preweaning and 3 days postweaning using a baby bottle. The effects on the weaning stress response were assessed through salivary cortisol measurements; behavioral tests such as the open field, novel object end elevated plus maze tests; and gene expression of HSD11B1, NR3C1 and IL1B in PBMCs, which was determined by RT-qPCR and normalized to GAPDH and UBB. To test the effect of the supplementations on weight, milk intake, gene expression, and behavior, a randomized block design was used with repeated measurements over time by the PROC MIXED procedure. Results and discussion: The effects on salivary cortisol were determined using the ratio between the morning and afternoon concentrations, considering the time before and after the weaning event. Principal component analysis (PCA) and Fisher's test were performed to evaluate the behavior test data. When comparing salivary cortisol concentrations between the pre- and postweaning periods, there was a difference (p < 0.05) between the supplementation groups in the afternoon period, suggesting that piglets fed asinine milk had lower afternoon cortisol concentrations postweaning than their counterparts. For the behavioral tests, the supplementations had no measurable effects. No difference was between groups pre- and postweaning for the expression of HSD11B2, which codes for an enzyme that breaks down cortisol. However, the expression of NR3C1, which encodes the glucocorticoid receptor, was significantly upregulated in piglets supplemented with cow milk (mean 1.245; p < 0.05). Conclusion: Asinine milk downregulated 1L1B gene expression, which codes for an inflammatory cytokine. In conclusion, these results suggest that supplementation with asinine milk may represent a strategy to diminish the damage associated with an early life event by modulating IL1B expression and reducing salivary cortisol levels in piglets undergoing weaning stress. Further transcriptomic and metabolomic studies may improve our understanding of the molecular pathways that mediate this systemic immune-mediated response.

Housing Conditions and a Challenge with Lipopolysaccharide on the Day of Estrus Can Influence Gene Expression of the Corpus Luteum in Gilts.

The corpus luteum (CL) is a temporary endocrine gland that plays a decisive role in the reproductive physiology of gilts. Recently, it has been suggested that exogenous factors may compromise the normal functioning of the CL. In the present study, we aimed to understand to what extent an acute and systemic challenge with lipopolysaccharide (LPS) on the day of estrus could compromise gene expression of gilts' CLs housed in different welfare conditions. For this, we housed 42 gilts in three different housing systems: crates, indoor group pens, and outdoor housing. Then, we challenged six females from each group with LPS and eight with saline (SAL) on the day of estrus. After slaughtering the gilts on the fifth day after the challenge, ovaries were collected for gene expression analysis by RT-qPCR. Housing system and LPS challenge did not have a significant interaction for any genes evaluated; thus, their effects were studied separately. We identified significant (p < 0.05) downregulation of the angiogenic genes VEGF and FTL1 among LPS-challenged animals. Meanwhile, we also observed upregulation of HSD3B1 gene among LPS-challenged animals. We found that STAR and LHCGR genes were differentially expressed depending on the housing system, which indicates that the environment may affect adaptation capabilities. Our results indicate that an acute health challenge on the estrus day alters CL gene expression; however, the role of the housing system remains uncertain.