The effect of mixing levels on aggression at weaning in piglets

Reducing Weaning Stress in Piglets by Pre-Weaning Socialization and Gradual Separation from the Sow: A Review

The weaning of pigs in most commercial pork production systems is an abrupt event performed at a fairly young age, i.e., mostly between 2.5 and 5 weeks of age. This practice induces a stress response, and its impact on behavior, performance and the gastrointestinal tract has been well described. Historically, there has been a focus on pre- and post-weaning nutritional strategies and post-weaning housing conditions and medication to improve production and reduce mortality after weaning. However, alternative pre-weaning housing and management systems that promote the development of natural social behaviors of piglets before weaning have recently received more attention. Co-mingling of non-littermates before weaning is a strategy that aims to initiate social interactions prior to weaning. The separation of the litter from the sow in the period leading up to weaning, termed intermittent suckling, aims to enhance the gradual separation from the sow. In addition, these practices encourage the young pig to learn explorative nutrient sourcing. Altogether, they may reduce weaning-associated stress. In this review, these strategies are defined, and their effects on behavior, performance, mortality, gastrointestinal function and immunocompetence are described. Though these strategies may be adapted to a commercial setting, it also becomes clear that many factors can contribute to the success of these strategies.

A computer vision image differential approach for automatic detection of aggressive behavior in pigs using deep learning

Pig aggression is a major problem facing the industry as it negatively affects both the welfare and the productivity of group-housed pigs. This study aimed to use a supervised deep learning (DL) approach based on a convolutional neural network (CNN) and image differential to automatically detect aggressive behaviors in pairs of pigs. Different pairs of unfamiliar piglets (N = 32) were placed into one of the two observation pens for 3 d, where they were video recorded each day for 1 h following mixing, resulting in 16 h of video recordings of which 1.25 h were selected for modeling. Four different approaches based on the number of frames skipped (1, 5, or 10 for Diff1, Diff5, and Diff10, respectively) and the amalgamation of multiple image differences into one (blended) were used to create four different datasets. Two CNN models were tested, with architectures based on the Visual Geometry Group (VGG) VGG-16 model architecture, consisting of convolutional layers, max-pooling layers, dense layers, and dropout layers. While both models had similar architectures, the second CNN model included stacked convolutional layers. Nine different sigmoid activation function thresholds between 0.1 and 1.0 were evaluated and a 0.5 threshold was selected to be used for testing. The stacked CNN model correctly predicted aggressive behaviors with the highest testing accuracy (0.79), precision (0.81), recall (0.77), and area under the curve (0.86) values. When analyzing the model recall for behavior subtypes prediction, mounting and mobile non-aggressive behaviors were the hardest to classify (recall = 0.63 and 0.75), while head biting, immobile, and parallel pressing were easy to classify (recall = 0.95, 0.94, and 0.91). Runtimes were also analyzed with the blended dataset, taking four times less time to train and validate than the Diff1, Diff5, and Diff10 datasets. Preprocessing time was reduced by up to 2.3 times in the blended dataset compared to the other datasets and, when combined with testing runtimes, it satisfied the requirements for real-time systems capable of detecting aggressive behavior in pairs of pigs. Overall, these results show that using a CNN and image differential-based deep learning approach can be an effective and computationally efficient technique to automatically detect aggressive behaviors in pigs.

Physiological response of weaned piglets to two transport durations observed in a Canadian commercial setting

Observational studies describing the impact of transport duration on weaned piglet welfare are limited. Current Canadian transport regulations are heavily informed by studies involving market hogs. Due to physiological differences between weaned piglets and market hogs, additional data on their response to transport are needed for age-specific evidence-based recommendations. A cohort study was conducted to describe and compare mortality, injury, weight change, hematological or biochemical changes in hydration, muscle injury and stress response observed in weaned piglets undergoing short duration (SD, <3 h), or long duration (LD, >30 h) commercial summertime transport events. Data collection on 440 of 11,434 transported piglets occurred the morning of the day before transport (T0), at arrival (T1) and approximately 3 to 4 d (78 to 93 h) after arrival at the nursery barn (T2). Low mortality occurred over all transport events (0.06%) with no association observed between transport duration and odds of death during transport (P = 0.62). The incidence of lameness between T0 and T1 was low (1.84% of the 435 focal piglets scored) with all lameness cases identified as mild in severity. Lesions on ears and skin were more prevalent than other injury types after transport (T1) and may have been related to mixing aggression associated with weaning rather than transport alone. LD piglets weighed 0.39 kg less than SD piglets at T1 (P < 0.01), but no difference in group weight was observed at T2 (P = 0.17). Hematological and biochemical differences were present between groups at T1. LD piglets had increased hematocrit levels compared with SD piglets (P = 0.01), suggesting increased body water losses. SD piglets showed greater levels of muscle injury compared with LD piglets including elevated aspartate aminotransferase (P < 0.01) and creatine kinase (P < 0.01). However, these parameters were within normal reference ranges for piglets of this age group. Indicators of physiological stress response including cortisol and neutrophil to lymphocyte ratios were elevated in SD piglets compared with LD piglets (P = 0.02 and P < 0.01, respectively). The results of this study demonstrate that both short and long transport durations can result in detectable physiological changes in weaned piglets. The overall impact of these durations on piglet welfare should be further explored by analyzing behavioral time budgets during and after transport.