Explore the vast range of titles available.

Changes in pig behaviours are a useful aid in detecting early signs of compromised health and welfare. In commercial settings, automatic detection of pig behaviours through visual imaging remains a challenge due to farm demanding conditions, e.g., occlusion of one pig from another. Here, two deep learning-based detector methods were developed to identify pig postures and drinking behaviours of group-housed pigs. We first tested the system ability to detect changes in these measures at group-level during routine management. We then demonstrated the ability of our automated methods to identify behaviours of individual animals with a mean average precision of 0.989 ± 0.009 , under a variety of settings. When the pig feeding regime was disrupted, we automatically detected the expected deviations from the daily feeding routine in standing, lateral lying and drinking behaviours. These experiments demonstrate that the method is capable of robustly and accurately monitoring individual pig behaviours under commercial conditions, without the need for additional sensors or individual pig identification, hence providing a scalable technology to improve the health and well-being of farm animals. The method has the potential to transform how livestock are monitored and address issues in livestock farming, such as targeted treatment of individuals with medication.

The review presents results of recent life cycle assessment studies aiming to quantify and improve the environmental performance of UK poultry production systems, including broiler meat, egg and turkey meat production. Although poultry production has been found to be relatively environmentally friendly compared with the production of other livestock commodities, it still contributes to environmental impacts, such as global warming, eutrophication and acidification. Amongst different sub-processes, feed production and transport contributes about 70 % to the global warming potential of poultry systems, whereas manure management contributes about 40–60 % to their eutrophication potential and acidification potential, respectively. All these impacts can be reduced by improving the feed efficiency, either by changing the birds through genetic selection or by making the feed more digestible (e.g. by using additives such as enzymes). However, although genetic selection has the potential to reduce the resources needed for broiler production (including feed consumption), the changing need of certain feed ingredients, most notably protein sources as a result of changes in bird requirements may limit the benefits of this strategy. The use of alternative feed ingredients, such as locally grown protein crops and agricultural by-products, as a replacement of South American grown soya, can potentially also lead to improvements in several environmental impact categories, as long as such feeding strategies have no negative effect on bird performance. Other management options, such as improving poultry housing and new strategies for manure management have also the potential to further improve the environmental sustainability of the poultry industries in Europe.

The objective of this study was to develop a fully automated multiple-cow real-time lameness detection system using a deep learning approach for cattle detection and pose estimation that could be deployed across dairy farms. Utilising computer vision and deep learning, the system can analyse simultaneously both the posture and gait of each cow within a camera field of view to a very high degree of accuracy (94–100%). Twenty-five video sequences containing 250 cows in varying degrees of lameness were recorded and independently scored by three accredited Agriculture and Horticulture Development Board (AHDB) mobility scorers using the AHDB dairy mobility scoring system to provide ground truth lameness data. These observers showed significant inter-observer reliability. Video sequences were broken down into their constituent frames and with a further 500 images downloaded from google, annotated with 15 anatomical points for each animal. A modified Mask-RCNN estimated the pose of each cow to output 5 key-points to determine back arching and 2 key-points to determine head position. Using the SORT (simple, online, and real-time tracking) algorithm, cows were tracked as they move through frames of the video sequence (i.e., in moving animals). All the features were combined using the CatBoost gradient boosting algorithm with accuracy being determined using threefold cross-validation including recursive feature elimination. Precision was assessed using Cohen’s kappa coefficient and assessments of precision and recall. This methodology was applied to cows with varying degrees of lameness (according to accredited scoring, n = 3) and demonstrated that some characteristics directly associated with lameness could be monitored simultaneously. By combining the algorithm results over time, more robust evaluation of individual cow lameness was obtained. The model showed high performance for predicting and matching the ground truth lameness data with the outputs of the algorithm. Overall, threefold lameness detection accuracy of 100% and a lameness severity classification accuracy of 94% respectively was achieved with a high degree of precision (Cohen’s kappa = 0.8782, precision = 0.8650 and recall = 0.9209).

We designed and evaluated an assumption-free, deep learning-based methodology for animal health monitoring, specifically for the early detection of respiratory disease in growing pigs based on environmental sensor data. Two recurrent neural networks (RNNs), each comprising gated recurrent units (GRUs), were used to create an autoencoder (GRU-AE) into which environmental data, collected from a variety of sensors, was processed to detect anomalies. An autoencoder is a type of network trained to reconstruct the patterns it is fed as input. By training the GRU-AE using environmental data that did not lead to an occurrence of respiratory disease, data that did not fit the pattern of “healthy environmental data” had a greater reconstruction error. All reconstruction errors were labelled as either normal or anomalous using threshold-based anomaly detection optimised with particle swarm optimisation (PSO), from which alerts are raised. The results from the GRU-AE method outperformed state-of-the-art techniques, raising alerts when such predictions deviated from the actual observations. The results show that a change in the environment can result in occurrences of pigs showing symptoms of respiratory disease within 1–7 days, meaning that there is a period of time during which their keepers can act to mitigate the negative effect of respiratory diseases, such as porcine reproductive and respiratory syndrome (PRRS), a common and destructive disease endemic in pigs.

Although pig systems start from a favourable baseline of environmental impact compared to other livestock systems, there is still scope to reduce their emissions and further mitigate associated impacts, especially in relation to nitrogen and phosphorous emissions. Key environmental impact hotspots of pig production systems are activities associated with feed production and manure management, as well as direct emissions (such as methane) from the animals and energy use. A major contributor to the environmental impacts associated with pig feed is the inclusion of soya in pig diets, especially since European pig systems rely heavily on soya imported from areas of the globe where crop production is associated with significant impacts of land use change, deforestation, carbon emissions, and loss of biodiversity. The “finishing” pig production stage contributes most to these environmental impacts, due to the amount of feed consumed, the efficiency with which feed is utilised, and the amount of manure produced during this stage. By definition therefore, any substantial improvements pig system environmental impact would arise from changes in feed production and manure management. In this paper, we consider potential solutions towards system environmental sustainability at these pig system components, as well as the bottlenecks that inhibit their effective implementation at the desired pace and magnitude. Examples include the quest for alternative protein sources to soya, the limits (perceived or real) to the genetic improvement of pigs, and the implementation of alternative manure management strategies, such as production of biogas through anaerobic digestion. The review identifies and discusses areas that future efforts can focus on, to further advance understanding around the potential sustainability benefits of modifications at various pig system components, and key sustainability trade-offs across the environment—economy—society pillars associated with synergistic and antagonistic effects when joint implementation of multiple solutions is considered. In this way, the review opens a discussion to facilitate the development of holistic decision support tools for pig farm management that account for interactions between the “feed * animal * manure” system components and trade-offs between sustainability priorities (e.g., environmental vs economic performance of pig system; welfare improvements vs environmental impacts).

Many members of the public and important stakeholders operating at the upper end of the food chain, may be unfamiliar with how food is produced, including within modern animal production systems. The intensification of production is becoming increasingly common in modern farming. However, intensive systems are particularly susceptible to production diseases, with potentially negative consequences for farm animal welfare (FAW). Previous research has demonstrated that the public are concerned about FAW, yet there has been little research into attitudes towards production diseases, and their approval of interventions to reduce these. This research explores the public's attitudes towards, and preferences for, FAW interventions in five European countries (Finland, Germany, Poland, Spain and the UK). An online survey was conducted for broilers (n = 789), layers (n = 790) and pigs (n = 751). Data were analysed by means of Kruskal-Wallis ANOVA, exploratory factor analysis and structural equation modelling. The results suggest that the public have concerns regarding intensive production systems, in relation to FAW, naturalness and the use of antibiotics. The most preferred interventions were the most \"proactive\" interventions, namely improved housing and hygiene measures. The least preferred interventions were medicine-based, which raised humane animal care and food safety concerns amongst respondents. The results highlighted the influence of the identified concerns, perceived risks and benefits on attitudes and subsequent behavioural intention, and the importance of supply chain stakeholders addressing these concerns in the subsequent communications with the public.

PurposeThe environmental impacts (EIs) of the global pig production sector are expected to increase with increasing global pork demand. Although the pig breeding industry has made significant progress over the last decades in reducing its EI, previous work has been unable to differentiate between the improvements made through management improvements from those caused by genetic change. Our study investigates the effect of altering genetic components of individual traits on the EI of pig systems.MethodsAn LCA model, with a functional unit of 1 kg live weight pig, was built simulating an intensive pig production system; inputs of feed and outputs of manure were adjusted according to genetic performance traits. Feed intake was simulated with an animal energy requirement model. A correlation matrix of the genetic variance and correlations of traits was pooled from data on commercial pig populations in the literature. Three sensitivity analyses were applied: one-at-a-time sensitivity analysis (OAT) used the genetic standard deviations, clusters-of-traits sensitivity analysis (COT) used the genetic standard deviations and clustering based on correlations, and the sensitivity index (SI) applied the full correlation matrix. Five EI categories were considered: global warming potential, terrestrial acidification potential, freshwater eutrophication potential, land use, and fossil resource scarcity.Results and discussionThe different EI categories showed similar behaviour for each trait in the sensitivity analyses. OAT showed up to 18% change in EI relative to baseline for energy maintenance and around 3% change in EI relative to baseline for most other traits. COT grouped traits into a grower/finisher cluster (up to 17% change relative to baseline), a reproductive cluster (up to 7% change relative to baseline), and a sow robustness cluster (up to 2% change relative to baseline), all clusters including negative correlations between traits. By including genetic correlations, the SI went from being influenced by maintenance, and finisher and gilt growth rate into solely being dominated by maintenancen and protein-to-lipid ratio responsible for above 0.8 and 0.35 of the variance in EI respectively.ConclusionsWe developed a novel methodology for evaluating EIs of changes in correlated genetic traits in pigs. We found it was essential to include correlations in the sensitivity analysis, since the local and global sensitivity analyses were not affected to the same extend by the same traits. Further, we found that finisher growth rate, body protein-to-lipid ratio, and energy maintenance could be important in reducing EI, but mortalities and sow robustness had little effect.

PurposeThe environmental and economic impacts of livestock production systems are typically assessed using global characterisation factors and data, even though several impact categories call for site-specific assessments. Here, we account for spatial variability by addressing potential interactions between geographic locality and the cost-effectiveness of farm investments that aim to reduce system environmental impact, using Danish pig production as a case-in-point.MethodsAn LCA-based, spatially explicit environmental abatement cost framework was developed to assess the cost-effectiveness of potential environmental abatement strategies. The framework was tested for Danish pig production in a “4 manure management × 4 geographic location” scenario analysis design. In addition to the baseline, the alternative manure management strategies were on-farm anaerobic digestion, slurry acidification and screw press slurry separation, implemented in an integrated pig farming system. The geographic locations differed in their proximity to Natura 2000 areas and in pig farming density. Eight different impact categories were assessed through an LCA using spatially explicit characterisation factors whenever possible, and annualised abatement potential was estimated for each manure management scenario and in each geographic location. We also estimated the financial performance for each scenario, through a discounted cash flow analysis at a whole-farm level.Results and discussionWe observed significant interactions between geographic location and system environmental and economic performance under baseline conditions. Significant location effects were also observed for the cost-effectiveness of all manure management strategies tested. Anaerobic digestion was the only “win–win” strategy that increased farm profits while reducing system environmental impact in two of the geographic cases: when implemented in a region of high pig farming density located near Natura 2000 and when implemented in a region of high pig farming density located far from Natura 2000 areas. Slurry acidification and slurry separation achieved sizeable abatement potential for impacts on ecosystem quality but incurred large additional costs in all geographic case studies considered, particularly when arable land was limited near the pig farm.ConclusionsAccounting for basic spatial characteristics within an environmental abatement cost framework had significant impact on the cost-effectiveness of on-farm investments for mitigation of system environmental impact. To the best of our knowledge, no studies to date have utilised such spatial characteristics within environmental abatement cost modelling of livestock farming systems. The presented framework has the potential to be further expanded using more detailed spatial, economic and geophysical data, which could ultimately improve decision-making regarding cost-effective investments that aim to improve the sustainability of livestock farming operations.

Background Conventional broilers are currently one of the most efficient protein converters. Although decades of progress in genetic selection and feed formulation have lead to high standards of efficient broiler production, still a lot of variability is found between farms and between successive flocks. The aim of this study was to investigate risk- and/or protective factors for poor health and performance in conventional broiler-farms in Europe by developing eight multivariable linear mixed models. Three different models were used to investigate mortality (overall, first week, after first week), three models for performance variables (growth, feed conversion, European production index) and two models were related to slaughterhouse data (i.e. dead on arrival and condemnation rate). Results Several factors related to management and housing were significantly associated with health and performance of broilers. The following factors were associated with increased mortality: floor quality, neonatal septicemia, ventilation type and other professional activities of the farmer. The factors associated with performance were chick sex, coccidiosis infections, necrotic enteritis, dysbacteriosis, light intensity adaptations, ventilation type, comparing daily flock results with previous flock results by farmer, daily check of feed and water system and type of feed. For dead on arrival three risk factors were identified i.e. daily growth, type of light adaptation and type of drinkers system. For condemnation rate seven risk factors were found, i.e. type of drinking system, daily growth, feed withdrawal time, type of ventilation, house size, septicemia after seven days and type of feed. Conclusions These results imply that a multifactorial approach is required with adaptations involving both improvements in management, housing, health programs and an increasing level of professionalism of the farmer in order to improve broiler performance and health.

We modified an automated method capable of quantifying behaviors which we then applied to the changes associated with the post-weaning transition in pigs. The method is data-driven and depends solely on video-captured image data without relying on sensors or additional pig markings. It was applied to video images generated from an experiment during which post-weaned piglets were subjected to treatments either containing or not containing in-feed antimicrobials (ZnO or antibiotics). These treatments were expected to affect piglet performance and health in the short-term by minimizing the risk from post-weaning enteric disorders, such as diarrhea. The method quantified total group feeding and drinking behaviors as well as posture (i.e., standing and non-standing) during the first week post-weaning, when the risk of post-weaning diarrhea is at its highest, by learning from the variations within each behavior using data manually annotated by a behavioral scientist. Automatically quantified changes in behavior were consistent with the effects of the absence of antimicrobials on pig performance and health, and manifested as reduced feed efficiency and looser feces. In these piglets both drinking and standing behaviors were increased during the first 6 days post-weaning. The correlation between fecal consistency and drinking behavior 6 days post weaning was relatively high, suggesting that these behaviors may have a diagnostic value. The presence or absence of in-feed antimicrobials had no effect on feeding behavior, which, however, increased over time. The approach developed here is capable of automatically monitoring several different behaviors of a group of pigs at the same time, and potentially this may be where its value as a diagnostic tool may lie.