Computer vision systems have the potential to provide automated, non-invasive monitoring of livestock animals, however, the lack of public datasets with well-defined targets and evaluation metrics presents a significant challenge for researchers. Consequently, existing solutions often focus on achieving task-specific objectives using relatively small, private datasets. This work introduces a new dataset and method for instance-level detection of multiple pigs in group-housed environments. The method uses a single fully-convolutional neural network to detect the location and orientation of each animal, where both body part locations and pairwise associations are represented in the image space. Accompanying this method is a new dataset containing 2000 annotated images with 24,842 individually annotated pigs from 17 different locations. The proposed method achieves over 99% precision and over 96% recall when detecting pigs in environments previously seen by the network during training. To evaluate the robustness of the trained network, it is also tested on environments and lighting conditions unseen in the training set, where it achieves 91% precision and 67% recall. The dataset is publicly available for download.

Tracking individual animals in a group setting is a exigent task for computer vision and animal science researchers. When the objective is months of uninterrupted tracking and the targeted animals lack discernible differences in their physical characteristics, this task introduces significant challenges. To address these challenges, a probabilistic tracking-by-detection method is proposed. The tracking method uses, as input, visible keypoints of individual animals provided by a fully-convolutional detector. Individual animals are also equipped with ear tags that are used by a classification network to assign unique identification to instances. The fixed cardinality of the targets is leveraged to create a continuous set of tracks and the forward-backward algorithm is used to assign ear-tag identification probabilities to each detected instance. Tracking achieves real-time performance on consumer-grade hardware, in part because it does not rely on complex, costly, graph-based optimizations. A publicly available, human-annotated dataset is introduced to evaluate tracking performance. This dataset contains 15 half-hour long videos of pigs with various ages/sizes, facility environments, and activity levels. Results demonstrate that the proposed method achieves an average precision and recall greater than 95% across the entire dataset. Analysis of the error events reveals environmental conditions and social interactions that are most likely to cause errors in real-world deployments.

Our recent research showed that antibody response to porcine reproductive and respiratory syndrome (PRRS), measured as sample-to-positive (S/P) ratio, is highly heritable and has a high genetic correlation with reproductive performance during a PRRS outbreak. Two major quantitative trait loci (QTL) on Sus scrofa chromosome 7 (SSC7; QTLMHC and QTL130) accounted for ~40 % of the genetic variance for S/P. Objectives of this study were to estimate genetic parameters for PRRS S/P in gilts during acclimation, identify regions associated with S/P, and evaluate the accuracy of genomic prediction of S/P across populations with different prevalences of PRRS and using different single nucleotide polymorphism (SNP) sets. Phenotypes and high-density SNP genotypes of female pigs from two datasets were used. The outbreak dataset included 607 animals from one multiplier herd, whereas the gilt acclimation (GA) dataset included data on 2364 replacement gilts from seven breeding companies placed on health-challenged farms. Genomic prediction was evaluated using GA for training and validation, and using GA for training and outbreak for validation. Predictions were based on SNPs across the genome (SNPAll), SNPs in one (SNPMHC and SNP130) or both (SNPSSC7) QTL, or SNPs outside the QTL (SNPRest). Heritability of S/P in the GA dataset increased with the proportion of PRRS-positive animals in the herd (from 0.28 to 0.47). Genomic prediction accuracies ranged from low to moderate. Average accuracies were highest when using only the 269 SNPs in both QTL regions (SNPSSC7, with accuracies of 0.39 and 0.31 for outbreak and GA validation datasets, respectively. Average accuracies for SNPALL, SNPMHC, SNP130, and SNPRest were, respectively, 0.26, 0.39, 0.21, and 0.05 for the outbreak, and 0.28, 0.25, 0.22, and 0.12, for the GA validation datasets. Moderate genomic prediction accuracies can be obtained for PRRS antibody response using SNPs located within two major QTL on SSC7, while the rest of the genome showed limited predictive ability. Results were obtained using data from multiple genetic sources and farms, which further strengthens these findings. Further research is needed to validate the use of S/P ratio as an indicator trait for reproductive performance during PRRS outbreaks.

The objectives were to determine the sensitivity, specificity, and cutoff values of a visual-based precision livestock technology (NUtrack), and determine the sensitivity and specificity of sickness score data collected with the live observation by trained human observers. At weaning, pigs (n = 192; gilts and barrows) were randomly assigned to one of twelve pens (16/pen) and treatments were randomly assigned to pens. Sham-pen pigs all received subcutaneous saline (3 mL). For LPS-pen pigs, all pigs received subcutaneous lipopolysaccharide (LPS; 300 μg/kg BW; E. coli O111:B4; in 3 mL of saline). For the last treatment, eight pigs were randomly assigned to receive LPS, and the other eight were sham (same methods as above; half-and-half pens). Human data from the day of the challenge presented high true positive and low false positive rates (88.5% sensitivity; 85.4% specificity; 0.871 Area Under Curve, AUC), however, these values declined when half-and-half pigs were scored (75% sensitivity; 65.5% specificity; 0.703 AUC). Precision technology measures had excellent AUC, sensitivity, and specificity for the first 72 h after treatment and AUC values were >0.970, regardless of pen treatment. These results indicate that precision technology has a greater potential for identifying pigs during a natural infectious disease event than trained professionals using timepoint sampling.

This study investigated the impact of an increased arginine (ARG) level in a boar diet on semen production, sperm quality, and seminal plasma proteome. Adult Nebraska Index Line boars were assigned to two groups, one receiving a control diet with 0.77% arginine (n = 4) and the other a high-arginine diet with 1.77% arginine (n = 5). Semen was collected twice a week over the whole experiment, including one week before, six weeks during, and six weeks after the supplementation. Parameters such as semen volume and concentration were assessed immediately after collection, alongside sperm motility and morphology. Centrifugation of raw semen samples yielded seminal plasma for a gel-based proteome analysis. The seminal plasma proteins were extracted, quantified, and separated via 2D gel electrophoresis, allowing protein identification through mass spectrometry. Data analysis involved two-way ANOVA for comparisons (p < 0.05). Results showed that arginine supplementation improved semen volume and total sperm counts, with averages of 21 ± 3 doses in the control group versus 24 ± 2 in the ARG group (p = 0.05). Although sperm motility and morphology remained unaffected (p > 0.05), dietary arginine upregulated ten proteins and downregulated two. In summary, increased dietary arginine did not significantly alter key parameters of semen output or sperm quality but significantly impacted seminal plasma proteome, warranting further research on sperm viability.