Animal-Based Indicators for On-Farm Welfare Assessment in Sheep

Investigating the use of machine learning algorithms to support risk-based animal welfare inspections of cattle and pig farms

In livestock production, animal-related data are often registered in specialised databases and are usually not interconnected, except for a common identifier. Analysis of combined datasets and the possible inclusion of third-party information can provide a more complete picture or reveal complex relationships. The aim of this study was to develop a risk index to predict farms with an increased likelihood for animal welfare violations, defined as non-compliance during on-farm welfare inspections. A data-driven approach was chosen for this purpose, focusing on the combination of existing Swiss government databases and registers. Individual animal-level data were aggregated at the herd level. Since data collection and availability were best for cattle and pigs, the focus was on these two livestock species. We present machine learning models that can be used as a tool to plan and optimise risk-based on-farm welfare inspections by proposing a consolidated list of priority holdings to be visited. The results of previous on-farm welfare inspections were used to calibrate a binary welfare index, which is the prediction goal. The risk index is based on proxy information, such as the participation in animal welfare programmes with structured housing and outdoor access, herd type and size, or animal movement data. Since transparency of the model is critical both for public acceptance of such a data-driven index and farm control planning, the Random Forest model, for which the decision process can be illustrated, was investigated in depth. Using historical inspection data with an overall low prevalence of violations of approximately 4% for both species, the developed index was able to predict violations with a sensitivity of 81.2 and 79.5% for cattle and pig farms, respectively. The study has shown that combining multiple and heterogeneous data sources improves the quality of the models. Furthermore, privacy-preserving methods are applied to a research environment to explore the available data before restricting the feature space to the most relevant. This study demonstrates that data-driven monitoring of livestock populations is already possible with the existing datasets and the models developed can be a useful tool to plan and conduct risk-based animal welfare inspection.

The Use of Machine Learning to Predict Prevalence of Subclinical Mastitis in Dairy Sheep Farms

The objective of the study was to develop a computational model with which predictions regarding the level of prevalence of mastitis in dairy sheep farms could be performed. Data for the construction of the model were obtained from a large Greece-wide field study with 111 farms. Unsupervised learning methodology was applied for clustering data into two clusters based on 18 variables (17 independent variables related to health management practices applied in farms, climatological data at the locations of the farms, and the level of prevalence of subclinical mastitis as the target value). The K-means tool showed the highest significance for the classification of farms into two clusters for the construction of the computational model: median (interquartile range) prevalence of subclinical mastitis among farms was 20.0% (interquartile range: 15.8%) and 30.0% (16.0%) (p = 0.002). Supervised learning tools were subsequently used to predict the level of prevalence of the infection: decision trees, k-NN, neural networks, and Support vector machines. For each of these, combinations of hyperparameters were employed; 83 models were produced, and 4150 assessments were made in total. A computational model obtained by means of Support vector machines (kernel: 'linear', regularization parameter C = 3) was selected. Thereafter, the model was assessed through the results of the prevalence of subclinical mastitis in 373 records from sheep flocks unrelated to the ones employed for the selection of the model; the model was used for evaluation of the correct classification of the data in each of 373 sets, each of which included a test (prediction) subset with one record that referred to the farm under assessment. The median prevalence of the infection in farms classified by the model in each of the two categories was 10.4% (5.5%) and 36.3% (9.7%) (p < 0.0001). The overall accuracy of the model for the results presented by the K-means tool was 94.1%; for the estimation of the level of prevalence (<25.0%/≥25.0%) in the farms, it was 96.3%. The findings of this study indicate that machine learning algorithms can be usefully employed in predicting the level of subclinical mastitis in dairy sheep farms. This can facilitate setting up appropriate health management measures for interventions in the farms.

Animal Welfare Investigation of Akkaraman Sheep Farms in Different Provinces of Türkiye

In this study, it was aimed to reveal the animal welfare levels in Akkaraman sheep breed in Türkiye. In this direction, welfare assessment was carried out at the farm level with the Animal Needs Index (ANI 35L/2000) method in a total of 71 Akkaraman sheep flock applications on animals were carried out on a total of 1525 sheep. According to the ANI score scale, the average score of all farms was determined as 39.52. In the study, welfare scores were found as 38.32, 41.47, and 38.78 in Çankırı, Çorum and Kırşehir provinces, respectively (p = 0.034); it was found as 39.70, 40.14, and 38.69 in small (≤100), medium (100-200) and large (>200) farms (p = 0.535), respectively. While the Famacha and Fecal scores of sheep were found to be low score in sheep raised in Çankırı than in sheep raised in other two cities, the differences were found to be statistically significant in both parameters (p = 0.007 and 0.021). As a result, it has been observed that having opportunity for animals to go out to yard and pasture has a positive effect on animal welfare.

Acute post-orchiectomy pain does not reduce alpha rams' interest in feed resources

Sheep pain is an animal welfare issue monitored based on behavioral responses, including appetite. Dominant (alpha) males have priority for accessing limited feed resources, however, the effects of pain on feed interest in members of a group with defined social hierarchy are unknown. Our objective was to investigate effects of acute post-orchiectomy pain on alpha rams' interest in accessing a limited feed resource. Eighteen rams were randomly housed in pens of 3 rams. After acclimation, the first 5-d (consecutive) battery of a behavior test was performed. In this test, 180 g of the regular diet concentrate was placed in a portable trough in the center of the pen; this feed was supplemental to the diet and represented a limited, albeit strongly preferable feed resource. Rams were filmed for 5 min after the feed introduction. Hierarchical levels (alpha, beta, and gamma) were defined based on the social hierarchical index according to higher initiator and lower receptor agonistic behaviors from the social network analyses. After 15 d, a second 5-d behavioral test battery was repeated. On the following day, alpha rams were castrated. Flunixin meglumine was given immediately before surgery and a final behavioral test was performed 8 h post-orchiectomy, concurrent with an expected peak in postoperative pain. For all recordings, the latency, frequency, and duration of time that each ram had its mouth inside the feed trough were recorded, and the Unesp-Botucatu sheep acute pain scale pain scale (USAPS) was applied. The social hierarchical index was highest in alpha rams, followed by beta and gamma. The pain scores were statistically equivalent across the 11 evaluation days for beta and gamma rams, whereas there was an increase in the final evaluation for alpha. There was no difference in latency, frequency, and duration between alpha, beta, and gamma rams across evaluations. We concluded that acute post-orchiectomy pain did not decrease alpha rams' interest in accessing limited feed. Routine feeding offers a valuable chance to detect pain-related behavior using the USAPS in rams. However, dominance may confound appetite-related behaviors in assessing acute pain, as alpha rams' interest in limited feed remained unaffected by the pain.

Development of a Parturition Detection System for Korean Native Black Goats

Korean Native Black Goats deliver mainly during the cold season. However, in winter, there is a high risk of stunted growth and mortality for their newborns. Therefore, we conducted this study to develop a KNBG parturition detection system that detects and provides managers with early notification of the signs of parturition. The KNBG parturition detection system consists of triaxial accelerometers, gateways, a server, and parturition detection alarm terminals. Then, two different data, the labor and non-labor data, were acquired and a Decision Tree algorithm was used to classify them. After classifying the labor and non-labor states, the sum of the labor status data was multiplied by the activity count value to enhance the classification accuracy. Finally, the Labor Pain Index (LPI) was derived. Based on the LPI, the optimal processing time window was determined to be 10 min, and the threshold value for labor classification was determined to be 14 240.92. The parturition detection rate was 82.4%, with 14 out of 17 parturitions successfully detected, and the average parturition detection time was 90.6 min before the actual parturition time of the first kid. The KNBG parturition detection system is expected to reduce the risk of stunted growth and mortality due to hypothermia in KNBG kids by detecting parturition 90.6 min before the parturition of the first kid, with a success rate of 82.4%, enabling parturition nursing.

Review of Rat (Rattus norvegicus), Mouse (Mus musculus), Guinea pig (Cavia porcellus), and Rabbit (Oryctolagus cuniculus) Indicators for Welfare Assessment

The monitoring and assessment of animals is important for their health and welfare. The appropriate selection of multiple, validated, and feasible welfare assessment indicators is required to effectively identify compromises or improvements to animal welfare. Animal welfare indicators can be animal or resource based. Indicators can be collated to form assessment tools (e.g., grimace scales) or animal welfare assessment models (e.g., 5 Domains) and frameworks (e.g., 5 Freedoms). The literature contains a wide variety of indicators, with both types needed for effective animal welfare assessment; however, there is yet to be an ideal constellation of indicators for animal-based welfare assessment in small mammals such as guinea pigs (Cavia Porcellus), mice (Mus musculus), rabbits (Oryctolagus cuniculus), and rats (Rattus norvegicus). A systematic review of grey and peer-reviewed literature was performed to determine the types of animal-based welfare indicators available to identify and assess animal health and welfare in these small mammals maintained across a wide variety of conditions. The available indicators were categorised and scored against a selection of criteria, including potential ease of use and costs. This review and analysis aim to provide the basis for further research into animal welfare indicators for these species. Future applications of this work may include improvements to animal welfare assessments or schemes, guiding better management, and implementing future strategies to enable better animal welfare.

Development of a data-driven method for assessing health and welfare in the most common livestock species in Switzerland: The Smart Animal Health project

Improving animal health and welfare in livestock systems depends on reliable proxies for assessment and monitoring. The aim of this project was to develop a novel method that relies on animal-based indicators and data-driven metrics for assessing health and welfare at farm level for the most common livestock species in Switzerland. Method development followed a uniform multi-stage process for each species. Scientific literature was systematically reviewed to identify potential health and welfare indicators for cattle, sheep, goats, pigs and poultry. Suitable indicators were applied in the field and compared with outcomes of the Welfare Quality® scores of a given farm. To identify farms at risk for violations of animal welfare regulations, several agricultural and animal health databases were interconnected and various supervised machine-learning techniques were applied to model the status of farms. Literature reviews identified a variety of indicators, some of which are well established, while others lack reliability or practicability, or still need further validation. Data quality and availability strongly varied among animal species, with most data available for dairy cows and pigs. Data-based indicators were almost exclusively limited to the categories “Animal health” and “Husbandry and feeding”. The assessment of “Appropriate behavior” and “Freedom from pain, suffering, harm and anxiety” depended largely on indicators that had to be assessed and monitored on-farm. The different machine-learning techniques used to identify farms for risk-based animal welfare inspections reached similar classification performances with sensitivities above 80%. Features with the highest predictive weights were: Participation in federal ecological and animal welfare programs, farm demographics and farmers' notification discipline for animal movements. A common method with individual sets of indicators for each species was developed. The results show that, depending on data availability for the individual animal categories, models based on proxy data can achieve high correlations with animal health and welfare assessed on-farm. Nevertheless, for sufficient validity, a combination of data-based indicators and on-farm assessments is currently required. For a broad implementation of the methods, alternatives to extensive manual on-farm assessments are needed, whereby smart farming technologies have great potential to support the assessment if the specific monitoring goals are defined.

The Importance of the Slaughterhouse in Surveilling Animal and Public Health: A Systematic Review

From the point of public health, the objective of the slaughterhouse is to guarantee the safety of meat in which meat inspection represent an essential tool to control animal diseases and guarantee the public health. The slaughterhouse can be used as surveillance center for livestock diseases. However, other aspects related with animal and human health, such as epidemiology and disease control in primary production, control of animal welfare on the farm, surveillance of zoonotic agents responsible for food poisoning, as well as surveillance and control of antimicrobial resistance, can be monitored. These controls should not be seen as a last defensive barrier but rather as a complement to the controls carried out on the farm. Regarding the control of diseases in livestock, scientific research is scarce and outdated, not taking advantage of the potential for disease control. Animal welfare in primary production and during transport can be monitored throughout ante-mortem and post-mortem inspection at the slaughterhouse, providing valuable individual data on animal welfare. Surveillance and research regarding antimicrobial resistance (AMR) at slaughterhouses is scarce, mainly in cattle, sheep, and goats. However, most of the zoonotic pathogens are sensitive to the antibiotics studied. Moreover, the prevalence at the slaughterhouse of zoonotic and foodborne agents seems to be low, but a lack of harmonization in terms of control and communication may lead to underestimate its real prevalence.

Positive Welfare Indicators in Dairy Animals

Nowadays, there is growing interest in positive animal welfare not only from the view of scientists but also from that of society. The consumer demands more sustainable livestock production, and animal welfare is an essential part of sustainability, so there is interest in incorporating positive welfare indicators into welfare assessment schemes and legislation. The aim of this review is to cite all the positive welfare indicators that have been proposed for dairy animals in theory or practice. In total, twenty-four indicators were retrieved. The most promising are exploration, access to pasture, comfort and resting, feeding, and behavioral synchronicity. Qualitative behavioral assessment (QBA), social affiliative behaviors, play, maternal care, ear postures, vocalizations, visible eye white, nasal temperature, anticipation, cognitive bias, laterality, and oxytocin have been also studied in dairy ruminants. QBA is the indicator that is most often used for the on-farm welfare assessment. Among all dairy animals, studies have been performed mostly on cattle, followed by sheep and goats, and finally buffaloes. The research on camel welfare is limited. Therefore, there is a need for further research and official assessment protocols for buffaloes and especially camels.