Emergency contingency plans for the U.S. swine industry, including depopulation protocols, that are essential to limit the spread of sudden outbreaks of infectious diseases is currently lacking. Although novel depopulation methods such as water-based foam (WBF) and high-expansion nitrogen foam (N2F) are being investigated, carbon dioxide gas (CO2) is currently the only American Veterinary Medical Association (AVMA)-preferred method capable of depopulation of groups of swine. The AVMA's assessment of depopulation methods evaluates efficacy, animal welfare implications, and caretaker physical and mental health, in addition to logistical aspects of equipment acquisition and use. These criteria are best gauged using input from individuals familiar with the operations within the swine industry. Therefore, the aim of this study was to describe swine industry stakeholders' perceptions of WBF, N2F and CO2 depopulation after a large-scale field demonstration. A survey was created based on the criteria outlined in the AVMA Depopulation Guidelines to determine respondent perceptions of each method. Swine industry stakeholders of various backgrounds (N=32) were recruited and invited to observe demonstrations of each method. Mixed linear regression models were built to investigate the association between survey question scores and depopulation method. Respondents varied in occupation, with 37.5 % (12/32) belonging to an academic institution or veterinary medical association, 21.9 % (7/32) to a pork organization, and 18.8 % (6/32) to state or federal regulatory agencies. The remaining 21.8 % (7/32) was a group consisting of one producer (3.1 %), one individual in swine industry retail (3.1 %), one veterinarian in a private practice setting (3.1 %) and four (12.5 %) who did not disclose their affiliation. Average experience (±SD) in the swine industry was 14.4 (±12.4) years, and 40.6 % (13/32) had previous experience of any type in swine depopulation. The overall method impression scores revealed that WBF and N2F were perceived as better options compared to CO2 (P < 0.001). WBF and N2F scored higher on pig distress mitigation, protecting emotional and psychological health of personnel, and equipment accessibility compared to CO2 (P < 0.001). Stakeholders without a clear affiliation provided higher scores compared those affiliated with regulatory agencies, pork associations or academic or veterinary institutions, regarding minimizing pig distress, method safety/accessibility, and overall method impression. Few demographic differences were observed, suggesting similar perceptions of the three depopulation methods during the demonstrations. This industry feedback is valuable for future considerations, method improvements and facilitation for possible implementations into future response plans.

Introduction Precision livestock farming (PLF) technology development has proliferated recently, but on-farm adoption has lagged. Understanding PLF stakeholders’ views, practical applications, limitations, and concerns are necessary to understand the factors influencing the adoption of PLF technology. Methods Using semi-structured online interviews, 12 influential stakeholders’ PLF views and lived experiences were investigated. A phenomenological paradigm was used to generate qualitative data that was analyzed using template analysis. Results We identified two unique groups of stakeholders, namely the PLF enthusiast and PLF cautious groups. The majority of the participants were well aware and had firsthand experience with the PLF technologies that are currently being used in the swine industry. Discussion We found that PLF technology was perceived to improve specificity in decision-making, better care for pigs, improve animal health and welfare, increase labor efficiency, and improve resource-use efficiency. Poor internet connectivity and the inability to use PLF data for decision-making due to needing to first complete daily on-farm tasks were considered key obstacles to its implementation. To integrate PLF technology into the existing barn environment, it is necessary to modify farm buildings and infrastructure and management protocols. Stakeholders’ main concerns with PLF technology included data privacy issues and the influence of PLF technology on human-animal relationships and farmers’ duty of care to the animals. In conclusion, stakeholders perceived camera systems for monitoring pig health and welfare and ensuring individual pig identification as a high priority in PLF development going forward.

Automated behavior analysis (ABA) strategies are being researched at a rapid rate to detect an array of behaviors across a range of species. There is growing optimism that soon ethologists will not have to manually decode hours (and hours) of animal behavior videos, but that instead computers will process them for us. However, before we assume ABA is ready for practical use, it is important to take a realistic look at exactly what ABA is being developed, the expertise being used to develop it, and the context in which these studies occur. Once we understand common pitfalls occurring during ABA development and identify limitations, we can construct robust ABA tools to achieve automated (ultimately even continuous and real time) analysis of behavioral data, allowing for more detailed or longer-term studies of behavior on larger numbers of animals than ever before. ABA is only as good as it is trained to be. A key starting point is having manually annotated data for model training and assessment. However, most ABA developers are not trained in ethology. Often no formal ethogram is developed and descriptions of target behaviors in ABA publications are limited or inaccurate. In addition, ABA is also frequently developed using small datasets, which lack sufficient variability in animal morphometrics, activities, camera viewpoints, and environmental features to be generalizable. Thus, ABA often needs to be further validated before being used satisfactorily on different populations or under other conditions, even for research purposes. Multidisciplinary teams of researchers including ethologists and ethicists as well as computer scientists, data scientists, and engineers are needed to help address problems when applying computer vision ABA to measure behavior. Reference datasets that can be used for behavior detection should be generated and shared that include image data, annotations, and baseline analyses for benchmarking. Also critical is the development of standards for creating such reference datasets and descriptions of best practices for methods for validating results from detection tools to ensure they are robust and generalizable. At present, only a handful of publicly available datasets exist that can be used for development of ABA tools. As we work to realize the promise of ABA (and subsequent precision livestock farming technologies) to detect animal behavior, a clear understanding of best practices, access to accurately annotated datasets, and networking among ethologists and ABA developers will increase our chances for rapid and robust successes.

An early ethics assessment method was used to evaluate sustainability goals and early findings from an automated body scanning technology for swine production. The project had twin goals of discovering potential pitfalls in the technology and exploring the applicability of the method, derived from the Ethical Matrix, as a tool to aid researchers in product design at very early stages in the research and development (R&D) process. This paper reports results on the second objective. Results of the evaluation workshop were coded and qualitatively analyzed. These results are reported and compared; the exercise is compared to the findings of other researchers using more traditional methods for ethical assessment on similar technologies, as well as standard social science methods for ascertaining economic sustainability and social acceptability of technological innovations. We conclude that the method has promise, especially for its applicability at very early stages in R&D, but that it does not substitute for analyses that occur at a much later stage in product or procedural development.

Three penetrating captive bolt (PCB) placements were tested on cadaver heads from swine with estimated body weight (BW) >200 kg (sows = 232.9 ± 4.1 kg; boars = 229.3 ± 2.6 kg). The objectives were to determine tissue depth, cross-sectional brain area, visible brain damage (BD), regions of BD, and bolt-brain contact; and determine relationships between external head dimensions and tissue depth at each placement. A Jarvis PAS-Type P 0.25R PCB with a Long Stunning Rod Nosepiece Assembly and 3.5 g power loads was used at the following placements on heads from 111 sows and 46 boars after storage at 2 to 4 °C for ~62 h before treatment: FRONTAL (F)-3.5 cm superior to the optic orbits at midline, TEMPORAL (T)-at the depression posterior to the lateral canthus of the eye within the plane between the lateral canthus and the base of the ear, or BEHIND EAR (BE)-directly caudal to the pinna of the ear on the same plane as the eyes and targeting the middle of the opposite eye. For sows, the bolt path was in the plane of the brain for 42/42 (100%, 95% confidence interval [CI]: 91.6% to 100.0%) F heads, 39/40 (97.5%, 95% CI: 86.8% to 99.9%) T heads, and 34/39 (87.5%, 95% CI: 72.6% to 95.7%) BE heads; for the heads that could reliably be assessed for BD damage was detected in 25/26 (96.2%, 95% CI: 80.4% to 99.9%) F heads, 24/35 (68.6%, 95% CI: 50.7% to 83.2%) T heads, and 5/40 (12.5%, 95% CI: 4.2% to 26.8%) BE heads. For boars, the bolt path was in the plane of the brain for 17/17 (100.0%, 95% CI: 80.5% to 100.0%) F heads, 18/18 (100.0%, 95% CI: 81.5% to 100.0%) T heads, and 14/14 (100.0%, 95% CI: 76.8% to 100.0%) BE heads; damage was detected in 11/12 (91.7%, 95% CI: 61.5% to 99.8%) F heads, 2/15 (13.3%, 95% CI: 1.7% to 40.5%) T heads, and 7/14 (50.0%, 95% CI: 23.0% to 77.0%) BE heads. Tissue depth was reported as mean ± standard error followed by 95% one-sided upper reference limit (URL). For sows, total tissue thickness was different (P < 0.05) between placements (F: 52.7 ± 1.0 mm, URL: 64.1 mm; T: 69.8 ± 1.4 mm, URL: 83.9 mm; BE: 89.3 ± 1.5 mm, URL: 103.4 mm). In boars, total tissue thickness was different (P < 0.05) between placements (F: 41.2 ± 2.1 mm, URL: 56.3 mm; T: 73.2 ± 1.5 mm, URL: 83.4 mm; BE: 90.9 ± 3.5 mm, URL: 113.5 mm). For swine > 200 kg BW, F placement may be more effective than T or BE due to less soft tissue thickness, which may reduce concussive force. The brain was within the plane of bolt travel for 100% of F heads with BD for 96.2% and 91.7% of F sow and boar heads, respectively.

The objective of this study was to contrast the soft tissue thickness, cranial thickness, total tissue thickness, and cross-sectional brain area from the common frontal captive bolt placement for the captive bolt euthanasia of swine with the alternative temporal and caudal to pinna placements. One hundred and fifty-seven cadaver heads from sows and boars with estimated body weights greater than 200 kg were collected from a regional slaughter establishment following electrical stunning and assigned to the FRONTAL, TEMPORAL, or CAUDAL to pinna captive bolt placement treatments after cooling at 2–4°C for approximately 64 h. In sows, soft tissue thickness was different (P ˂ 0.0001) between the three placements (FRONTAL: 13.9±1.1 mm, TEMPORAL: 45.93±1.1 mm, CAUDAL TO PINNA: 53.8±1.1 mm), cranial thickness was different (P ˂ 0.0001) between the three placements (FRONTAL: 47.1±1.4 mm, TEMPORAL: 17.6±1.4 mm, CAUDAL TO PINNA: 30.2±1.4 mm), total tissue thickness was different (P < 0.0001) between the three placements (FRONTAL: 61.03±1.4 mm, TEMPORAL: 63.49±1.4 mm, CAUDAL TO PINNA: 84.05±1.4 mm), and cross-sectional brain area was different (P < 0.0001) between the three placements (FRONTAL: 4509.0±238.0 mm2, TEMPORAL: 1964.4±238.0 mm2, CAUDAL TO PINNA: 2767.5±238.0 mm2). In boars, soft tissue thickness was different (P < 0.0001) between the three placements (FRONTAL: 12.9±1.7mm, TEMPORAL: 45.3±1.7 mm, CAUDAL TO PINNA: 54.7±1.7 mm), cranial thickness was different (P = 0.0193) between the FRONTAL and TEMPORAL treatments (FRONTAL: 34.8±3.2mm, TEMPORAL: 22.1±3.2 mm, CAUDAL TO PINNA: 31.7±3.2 mm), total tissue thickness was different (P < 0.0001) between the three placements (FRONTAL: 47.7±3.2 mm, TEMPORAL: 67.4±3.2 mm, CAUDAL TO PINNA: 86.4±3.2mm), and cross-sectional brain area was different (P < 0.0001) between the three placements (FRONTAL: 4031.9±153.2mm2, TEMPORAL: 1241.8±153.2 mm2, CAUDAL TO PINNA: 2467.5±153.2 mm2). Overall, the preliminary data indicated that the FRONTAL placement appears to have the greatest likelihood for successful euthanasia and may present less risk than the alternative TEMPORAL or CAUDAL TO PINNA placements.

This chapter describes the application of technology in pig farming, including understand how computer systems learn, the different types of sensors to monitor health and behaviour, the basic operation of precision livestock farming (PLF) and future possibilities. Precision livestock farming (PLF) will use both electronic sensors and computerized machine learning to monitor the condition of pigs. Some of the animal-based variables that can be monitored are tail biting, coughing, activity, lameness, fighting, and body weight. To monitor the animals, some of the sensors that can be used are cameras, microphones, and temperature sensors. These sensors do not have to be attached to the animals. Computer algorithms have been developed that can recognize the faces of individual pigs. Other types of sensors, such as accelerometers for detecting lameness, can be attached to ear tags. Computerized artificial intelligence (AI) systems are continually improving. A system has to be able to accurately locate a lame animal without providing too many false positives.

Precision livestock farming uses artificial intelligence to individually monitor livestock activity and health. Tracking individuals over time can reveal health indicators that correlate with productivity and longevity. For instance, locomotion patterns observed in lame pigs have been shown to correlate with poor animal welfare and productivity. Kinematic analysis of pigs using pose estimates provides a means of assessing locomotion. New dense depth sensors have potential to achieve full 3D pose estimation and tracking. However, the lack of annotated dense depth datasets has limited use of these sensors in detecting animal pose. Current annotation methods rely on human labeling, but identifying hip and shoulder locations is difficult for pigs with few prominent features, and is especially difficult in depth images as these lack albedo texture. This work proposes a solution to quickly generate high accuracy pig landmark annotations for depth-based pose estimation. We propose Depth-Infrared Annotation Transfer (DIAT), an approach that semi-automatically finds, identifies, and tracks marks visible in infrared, and transfers these labels to depth images. As a result, we are able to train a precise pig pose detector that operates on depth images.

The burgeoning research and applications of technological advances are launching the development of precision livestock farming. Through sensors (cameras, microphones and accelerometers), images, sounds and movements are combined with algorithms to non-invasively monitor animals to detect their welfare and predict productivity. In turn, this remote monitoring of livestock can provide quantitative and early alerts to situations of poor welfare requiring the stockperson's attention. While swine practitioners' skills include translation of pig data entry into pig health and well-being indices, many do not yet have enough familiarity to advise their clients on the adoption of precision livestock farming practices. This review, intended for swine veterinarians and specialists, (1) includes an introduction to algorithms and machine learning, (2) summarizes current literature on relevant sensors and sensor network systems, and drawing from industry pig welfare audit criteria, (3) explains how these applications can be used to improve swine welfare and meet current pork production stakeholder expectations. Swine practitioners, by virtue of their animal and client advocacy roles, interpretation of benchmarking data, and stewardship in regulatory and traceability programs, can play a broader role as advisors in the transfer of precision livestock farming technology, and its implications to their clients.

Acute outbreaks of respiratory disease in swine at agricultural fairs in Michigan, USA, in 2015 raised concern for potential human exposure to influenza A virus. Testing ruled out influenza A virus and identified porcine hemagglutinating encephalomyelitis virus as the cause of influenza-like illness in the affected swine.

Abstract This review summarizes the effects of ractopamine hydrochloride (RAC) dose (5, 7.5, 10, and 20 mg/kg) on market weight pig welfare indicators. Ractopamine hydrochloride (trade name Paylean) is a β-adrenergic agonist that was initially approved in the U.S. in 1999 at doses of 5 to 20 mg/kg to improve feed efficiency and carcass leanness. However, anecdotal reports suggested that RAC increased the rate of non-ambulatory (fatigued and injured) pigs at U.S. packing plants. This led to the addition of a caution statement to the Paylean label, and a series of research studies investigating the effects of RAC on pig welfare. Early research indicated that: (1) regardless of RAC administration, fatigued (non-ambulatory, non-injured) pigs are in a state of metabolic acidosis; (2) aggressive handling increases stress responsiveness at 20 mg/kg RAC, while 5 mg/kg reduces stress responsiveness to aggressive handling. Given this information, dosage range for Paylean was changed in 2006 to 5 to 10 mg/kg in market weight pigs. Subsequent research on RAC demonstrated that: (1) RAC has minimal effects on mortality, lameness, and home pen behavior; (2) RAC fed pigs demonstrated inconsistent prevalence and intensity of aggressive behaviors; (3) RAC fed pigs may be more difficult to handle at doses above 5 mg/kg; and (4) RAC fed pigs may have increased stress responsiveness and higher rates of non-ambulatory pigs when subjected to aggressive handling, especially when 20 mg/kg of RAC is fed.

Transport losses (dead and nonambulatory pigs) present animal welfare, legal, and economic challenges to the US swine industry. The objectives of this review are to explore 1) the historical perspective of transport losses; 2) the incidence and economic implications of transport losses; and 3) the symptoms and metabolic characteristics of fatigued pigs. In 1933 and 1934, the incidence of dead and nonambulatory pigs was reported to be 0.08 and 0.16%, respectively. More recently, 23 commercial field trials (n = 6,660,569 pigs) were summarized and the frequency of dead pigs, nonambulatory pigs, and total transport losses at the processing plant were 0.25, 0.44, and 0.69% respectively. In 2006, total economic losses associated with these transport losses were estimated to cost the US pork industry approximately $46 million. Furthermore, 0.37 and 0.05% of the nonambulatory pigs were classified as either fatigued (nonambulatory, noninjured) or injured, respectively, in 18 of these trials (n = 4,966,419 pigs). Fatigued pigs display signs of acute stress (open-mouth breathing, skin discoloration, muscle tremors) and are in a metabolic state of acidosis, characterized by low blood pH and high blood lactate concentrations; however, the majority of fatigued pigs will recover with rest. Transport losses are a multifactorial problem consisting of people, pig, facility design, management, transportation, processing plant, and environmental factors, and, because of these multiple factors, continued research efforts are needed to understand how each of the factors and the relationships among factors affect the well-being of the pig during the marketing process.