Symposium review: Precision technologies for dairy calves and management applications

Early detection of bovine respiratory disease in pre-weaned dairy calves using sensor based feeding, movement, and social behavioural data

Previous research shows that feeding and activity behaviours in combination with machine learning algorithms has the potential to predict the onset of bovine respiratory disease (BRD). This study used 229 novel and previously researched feeding, movement, and social behavioural features with machine learning classification algorithms to predict BRD events in pre-weaned calves. Data for 172 group housed calves were collected using automatic milk feeding machines and ultrawideband location sensors. Health assessments were carried out twice weekly using a modified Wisconsin scoring system and calves were classified as sick if they had a Wisconsin score of five or above and/or a rectal temperature of 39.5 °C or higher. A gradient boosting machine classification algorithm produced moderate to high performance: accuracy (0.773), precision (0.776), sensitivity (0.625), specificity (0.872), and F1-score (0.689). The most important 30 features were 40% feeding, 50% movement, and 10% social behavioural features. Movement behaviours, specifically the distance walked per day, were most important for model prediction, whereas feeding and social features aided in the model's prediction minimally. These results highlighting the predictive potential in this area but the need for further improvement before behavioural changes can be used to reliably predict the onset of BRD in pre-weaned calves.

A qualitative survey approach to investigating beef and dairy veterinarians' needs in relation to technologies on farms

Globally, farmers are being increasingly encouraged to use technologies. Consequently, veterinarians often use farm data and technologies to provide farmers with advice. Yet very few studies have sought to understand veterinarians' perceptions of data and technologies on farms. The aim of this study was to understand veterinarians' experiences and opinions on data and technology on beef and dairy farms. An online qualitative survey was conducted with a convenience sample of 36 and 24 veterinarians from the United Kingdom and Ireland, respectively. The data were analysed using reflexive thematic analysis to generate four themes: (1) Improving veterinary advice through data; (2) Ensuring stock person skills are retained; (3) Longevity of technology; and (4) Solving social problems on farms. We show that technologies and data can make veterinarians feel more confident in the advice they give to farmers. However, the quality and quantity of data collected on cattle farms were highly variable. Furthermore, veterinarians were concerned that farmers can become over-reliant on technologies by not using their stockperson skills. As herd sizes increase, technologies can help to improve working conditions on farms with multiple employees of various skillsets. Veterinarians would like innovations that can help them to demonstrate their competence, influence farmers' behaviour, and ensure sustainability of the beef and dairy industries.

Review: Opportunities and challenges for the genetic selection of dairy calf disease traits

Interest in dairy cow health continues to grow as we better understand health's relationship with production potential and animal welfare. Over the past decade, efforts have been made to incorporate health traits into national genetic evaluations. However, they have focused on the mature cow, with calf health largely being neglected. Diarrhoea and respiratory disease comprise the main illnesses with regard to calf health. Conventional methods to control calf disease involve early separation of calves from the dam and housing calves individually. However, public concern regarding these methods, and growing evidence that these methods may negatively impact calf development, mean the dairy industry may move away from these practices. Genetic selection may be a promising tool to address these major disease issues. In this review, we examined current literature for enhancing calf health through genetics and discussed alternative approaches to improve calf health via the use of epidemiological modelling approaches, and the potential of indirectly selecting for improved calf health through improving colostrum quality. Heritability estimates on the observed scale for diarrhoea ranged from 0.03 to 0.20, while for respiratory disease, estimates ranged from 0.02 to 0.24. The breadth in these ranges is due, at least in part, to differences in disease prevalence, population structure, data editing and models, as well as data collection practices, which should be all considered when comparing literature values. Incorporation of epidemiological theory into quantitative genetics provides an opportunity to better determine the level of genetic variation in disease traits, as it accounts for disease transmission among contemporaries. Colostrum intake is a major determinant of whether a calf develops either respiratory disease or diarrhoea. Colostrum traits have the advantage of being measured and reported on a continuous scale, which removes the issues classically associated with binary disease traits. Overall, genetic selection for improved calf health is feasible. However, to ensure the maximum response, first steps by any industry members should focus efforts on standardising recording practices and encouragement of uploading information to genetic evaluation centres through herd management software, as high-quality phenotypes are the backbone of any successful breeding programme.

A living lab approach to understanding dairy farmers' needs of technologies and data to improve herd health: Focus groups from 6 European countries

For successful development and adoption of technology on dairy farms, farmers need to be included in the innovation process. However, the design of agricultural technologies usually takes a top-down approach with little involvement of end-users at the early stages. Living Labs offer a methodology that involve end-users throughout the development process and emphasize the importance of understanding users' needs. Currently, exploration of dairy farmers' needs of technologies has been limited to specific types of technology (e.g., smartphone apps) and adult cattle. The aim of this study was to use a Living Lab approach to identify dairy farmers' needs of data and technologies to improve herd health and inform innovation development. Eighteen focus groups were conducted with, in total, 80 dairy farmers from Belgium, Ireland, the Netherlands, Norway, Sweden, and the UK. Data were analyzed using Template Analysis and 6 themes were generated which represented the fundamental needs of autonomy, comfort, competence, community and relatedness, purpose, and security. Farmers favored technologies that provided them with convenience, facilitated their knowledge and understanding of problems on farm, and allowed them to be self-reliant. Issues with data sharing and accessibility, and usability of software were barriers to technology use. Furthermore, farmers were facing problems around recruitment and management of labor and needed ways to reduce stress. Controlling aspects of the barn environment, such as air quality, hygiene, and stocking density, was a particular concern in relation to youngstock management. In conclusion, the findings suggest that developers of farm technologies may want to include farmers in the design process to ensure a positive user experience and improve accessibility. The needs identified in this study can be used as a framework when designing farm technologies to strengthen need satisfaction and reduce any potential harm toward needs.

How can better calf care be realized on dairy farms? A qualitative interview study of veterinarians and farmers

Improving health and welfare outcomes for replacement and surplus dairy calves is important for the sustainability of the dairy industry. Dairy farmers and veterinarians hold expertise in calf management and have valuable perspectives on how to practically motivate improvements. The objective of this study was to determine strategies that could improve the care calves receive on dairy farms from the perspective of dairy farmers and their herd veterinarians. Two veterinary clinics specializing in dairy practice in British Columbia, Canada, and 21 of their client dairy farms participated in the project. Following a meeting in which calf colostrum management was discussed between farmers and their herd veterinarian, participant interviews were conducted. Separate interviews were conducted for the farmers (n = 27 farmers from 21 farms) and their herd veterinarians (n = 7, with 1 to 5 farms that each vet worked with enrolled in the study) using tailored semi-structured question guides. Interviews (n = 42) were transcribed and coded following inductive thematic analysis methodology. The themes identified included strategies for farmers, veterinarians, and calf buyers, as well as contexts that influenced the dairy farmers' internal motivation to provide good calf care. Results indicated that farmers could optimize their calf management through fostering engagement of calf care personnel or by enlisting technology. Veterinarians could provide support to farms by being actively involved in calf monitoring, assisting in developing operating protocols, and setting goals, and especially by using farm-specific data to guide their management recommendations. Calf buyers could communicate with and provide accountability to farmers and improve their purchasing strategies to encourage farms to raise more vigorous surplus calves. Farmers' personal values, social networks, and relationships with different dairy industry stakeholders influenced their concern about the standards of their calf care practices. These findings provide guidance on how dairy farmers could achieve or be prompted to achieve improvements in their calf care practices.

Strategies for Bovine Respiratory Disease (BRD) Diagnosis and Prognosis: A Comprehensive Overview

Despite significant advances in vaccination strategies and antibiotic therapy, bovine respiratory disease (BRD) continues to be the leading disease affecting the global cattle industry. The etiology of BRD is complex, often involving multiple microbial agents, which lead to intricate interactions between the host immune system and pathogens during various beef production stages. These interactions present environmental, social, and geographical challenges. Accurate diagnosis is essential for effective disease management. Nevertheless, correct identification of BRD cases remains a daunting challenge for animal health technicians in feedlots. In response to current regulations, there is a growing interest in refining clinical diagnoses of BRD to curb the overuse of antimicrobials. This shift marks a pivotal first step toward establishing a structured diagnostic framework for this disease. This review article provides an update on recent developments and future perspectives in clinical diagnostics and prognostic techniques for BRD, assessing their benefits and limitations. The methods discussed include the evaluation of clinical signs and animal behavior, biomarker analysis, molecular diagnostics, ultrasound imaging, and prognostic modeling. While some techniques show promise as standalone diagnostics, it is likely that a multifaceted approach-leveraging a combination of these methods-will yield the most accurate diagnosis of BRD.

Analyzing Cattle Activity Patterns with Ear Tag Accelerometer Data

In this study, we equip two breeds of cattle located in tropical and temperate climates with smart ear tags containing triaxial accelerometers to measure their activity levels across different time periods. We produce activity profiles when measured by each of four statistical features, the mean, median, standard deviation, and median absolute deviation of the Euclidean norm of either unfiltered or high-pass-filtered accelerometer readings over five-minute windows. We then aggregate the values from the 5 min windows into hourly or daily (24 h) totals to produce activity profiles for animals kept in each of the test environments. To gain a better understanding of the variation between the peak and nadir activity levels within a 24 h period, we divide each day into multiple equal-length intervals, which can range from 2 to 96 intervals. We then calculate a statistical measure, called daily differential activity (DDA), by computing the differences in feature values for each interval pair. Our findings demonstrate that patterns within the activity profile are more clearly visualised from readings that have been subject to high-pass filtering and that the median of the acceleration vector norm is the most reliable feature for characterising activity and calculating the DDA measure. The underlying causes for these differences remain elusive and is likely attributable to environmental factors, cattle breeds, or management practices. Activity profiles produced from the standard deviation (a feature routinely applied to the quantification of activity level) showed less uniformity between animals and larger variation in values overall. Assessing activity using ear tag accelerometers holds promise for monitoring animal health and welfare. However, optimal results may only be attainable when true diurnal patterns are detected and accounted for.

Association between a pyroelectric infrared sensor monitoring system and a 3-dimensional accelerometer to assess movement in preweaning dairy calves

The objective of this study was to correlate movement assessed by a pyroelectric infrared sensor system in preweaning dairy calves with lying and standing time assessed by a 3D accelerometer considering the temperature-humidity index (THI). A total of 35 dairy calves (1-7 d of age) were enrolled in the study and 20 calves were included in the final analyses. The lying and standing time of the calves was monitored with a 3D accelerometer (Hobo Pendant G Data Logger, Onset Computer Corporation, USA), which was used as the gold standard reference. The infrared sensor monitoring system (IMS; Calf Monitoring System, Futuro Farming GmbH, Germany) was fixed to the fence of the calf hutch within the calf's reach. Temperature-humidity was monitored with 2 validated THI sensors inside and on outside of each calf hutch. Additionally, one THI sensor was located near the calf hutches. The observation period lasted 14 consecutive days. The average standing time assessed by the 3D accelerometer was 13.4 ± 12.7 (mean ± standard deviation) min/h and the average lying time was 46.6 (±12.7) min/h. The median (25th percentile; 75th percentile) number of movements measured by the IMS was 360 (60; 919) movements per hour. Number of movements per hour measured by the IMS was compared with data obtained with a validated 3D accelerometer. The Pearson correlation coefficient between both standing and lying time and the number of movements was r = 0.85 and r = -0.85, respectively. The Pearson correlation coefficients were only slightly influenced by THI (low THI [<68]: r = 0.86; medium THI [68-72]: r = 0.85; high THI [>72]: r = 0.81). Our data show that the number of movements of dairy calves measured by IMS were highly correlated with the chosen gold standard reference method. High THI slightly affects the measurement accuracy of IMS.

The Effect of Feeding a Total Mixed Ration with an ad libitum or Restricted Pelleted Starter on Growth Performance, Rumination Behavior, Blood Metabolites, and Rumen Fermentation in Weaning Holstein Dairy Calves

The aim of the current study was to evaluate the effect of the starter restriction and of the ad libitum TMR (total mixed ration) inclusion on intake, growth performance, rumination time (RT), and health condition of Holstein dairy calves during weaning. We randomly assigned thirty female Holstein calves (with an average weight of 38.5 ± 1.96 kg at birth) to one of three treatments. From 21 days of age, the calves were fed one of three treatments as follows: a control diet (CTR) with an ad libitum calf starter but without TMR; Treatment 1 diet (TRT1) with both an ad libitum calf starter and ad libitum TMR; Treatment 2 diet (TRT2) with ad libitum TMR and a restricted amount of a calf starter (50% of the intake recorder in the control group day by day). Calves in the TRT2 group, between 56 and 63 days of age, had a lower body weight (80.1; 79.5; 75.6 kg for the CTR, TRT1, and TRT2 groups, respectively) compared with CTR and TRT1 calves. This outcome is ascribed to the average daily gain (0.759; 0.913; 0.508 kg/day for the CTR, TRT1, and TRT2 groups, respectively), resulting also in TRT2 being lower than CTR or TRT1 calves. The inclusion of ad libitum TMR increased the rumination time, especially after weaning (15.28 min/h, 18.38 min/h, and 18.95 min/h for the CTR, TRT1, and TRT2 groups, respectively). Concerning the rumen metabolism and inflammometabolic response, overall, no differences were observed between the three dietary treatments. In conclusion, the results indicated that a TMR could partially replace a calf starter in weaning dairy calves, since neither growth performance nor health status were impaired. In addition, providing TMR (with or without concentrate restriction) led to a better rumen development and likely a better rumen fermentation efficiency in post-weaning.

Exploring the data divide through a social practice lens: A qualitative study of UK cattle farmers

Appropriate management decisions are key for sustainable and profitable beef and dairy farming. Data-driven technologies aim to provide information which can improve farmers' decision-making practices. However, data-driven technologies have resulted in the emergence of a "data divide", in which there is a gap between the generation and use of data. Our study aims to further understand the data divide by drawing on social practice theory to recognise the emergence, linkages, and reproduction of youngstock data practices on cattle farms in the UK. Eight focus groups with fifteen beef and nineteen dairy farmers were completed. The topics of discussion included data use, technology use, disease management in youngstock, and future goals for their farm. The transcribed data were analysed using reflexive thematic analysis with a social practice lens. Social practice theory uses practices as the unit of analysis, rather than focusing on individual behaviours. Practices are formed of three elements: meaning (e.g., beliefs), materials (e.g., objects), and competencies (e.g., skills) and are connected in time and space. We conceptualised the data divide as a disconnection of data collection practices and data use and interpretation practices. Consequently, we were able to generate five themes that represent these breaks in connection.Our findings suggest that a data divide exists because of meanings that de-stabilise practices, tensions in farmers' competencies to perform practices, spatial and temporal disconnects, and lack of forms of feedback on data practices. The data preparation practice, where farmers had to merge different data sources or type up handwritten data, had negative meanings attached to it and was therefore sometimes not performed. Farmers tended to associate data and technology practices with larger dairy farms, which could restrict beef and small-scale dairy farms from performing these practices. Some farmers suggested that they lacked the skills to use technologies and struggled to transform their data into meaningful outputs. Data preparation and data use and interpretation practices were often tied to an office space because of the required infrastructure, but farmers preferred to spend time outdoors and with their animals. There appeared to be no normalisation of what data should be collected or what data should be analysed, which made it difficult for farmers to benchmark their progress. Some farmers did not have access to discussion groups or veterinarians who were interested in data and therefore could not get feedback on their data practices.These results suggest that the data divide exists because of three types of disconnect: a disconnect between elements within a practice because of tensions in competencies or negative meanings to perform a practice; a disconnect between practices because of temporal or spatial differences; and a break in the reproduction of practices because of lack of feedback on their practices. Data use on farms can be improved through transformation of practices by ensuring farmers have input in the design of technologies so that they align with their values and competencies.

Association between Prepartum Alerts Generated Using a Commercial Monitoring System and Health and Production Outcomes in Multiparous Dairy Cows in Five UK Herds

Identifying cows that are at greater risk for disease prior to calving would be a valuable addition to transition management. Prior to the commercial release of software features in an automated behavioral monitoring system, designed to identify cows in the dry period at greater risk of disease postpartum, a retrospective analysis was carried out in five dairy herds to evaluate whether the software could identify prepartum cows that subsequently received health treatments postpartum and whether prepartum alerts (transition alerts) are associated with a reduction in milk production in the subsequent lactation. Herd management and production records were analyzed for cows receiving treatment in the first 21 d of lactation (days in milk, DIM) for clinical mastitis, reproductive tract disease (metritis, retained fetal membranes), metabolic disease (hypocalcemia, ketosis and displaced abomasum) and for cows exiting the herd by 60 DIM. Data was gathered for 986 cows, 382 (38.7%) of which received a transition alert and 604 (61.3%) that did not. During the first 21 DIM 312 (31.6%) cows went on to receive a disease treatment, of these 51.9% (n = 162/312) were transition alert cows and 48.1% (n = 150/312) non-transition alert cows, while 8.6% (n = 33/382) alert cows exited the herd by 60 DIM compared to 4.8% (n = 29/604) of cows that did not receive an alert. A cow receiving a transition alert (OR = 1.76, 95% confidence interval (CI) = 1.27-2.44) and increasing parity (OR = 2.03, 95% CI = 1.44-2.86) were both associated with increased risk of receiving a disease treatment in the first 21 DIM. The occurrence of a transition alert was negatively associated with both week 4 milk yield (daily average yield in fourth week of lactation) and predicted 305 d yield. Transition alerts correctly predicted 62.5% (95% CI: 59.3-65.5) of treatments with a sensitivity of 42.4% (95% CI: 37.4-45.5) and a specificity of 75.2% (95% CI: 71.5-78.6). Associations were identified between postpartum health and production outcomes and prepartum behavioral measures from an automated activity monitoring system.

Intrinsic calf factors associated with the behavior of healthy pre-weaned group-housed dairy-bred calves

Technology-derived behaviors are researched for disease detection in artificially-reared calves. Whilst existing studies demonstrate differences in behaviors between healthy and diseased calves, intrinsic calf factors (e.g., sex and birthweight) that may affect these behaviors have received little systematic study. This study aimed to understand the impact of a range of calf factors on milk feeding and activity variables of dairy-bred calves. Calves were group-housed from ~7 days to 39 days of age. Seven liters of milk replacer was available daily from an automatic milk feeder, which recorded feeding behaviors and live-weight. Calves were health scored daily and a tri-axial accelerometer used to record activity variables. Healthy calves were selected by excluding data collected 3 days either side of a poor health score or a treatment event. Thirty-one calves with 10 days each were analyzed. Mixed models were used to identify which of live-weight, age, sex, season of birth, age of inclusion into the group, dam parity, birthweight, and sire breed type (beef or dairy), had a significant influence on milk feeding and activity variables. Heavier calves visited the milk machine more frequently for shorter visits, drank faster and were more likely to drink their daily milk allowance than lighter calves. Older calves had a shorter mean standing bout length and were less active than younger calves. Calves born in summer had a longer daily lying time, performed more lying and standing bouts/day and had shorter mean standing bouts than those born in autumn or winter. Male calves had a longer mean lying bout length, drank more slowly and were less likely to consume their daily milk allowance than their female counterparts. Calves that were born heavier had fewer lying and standing bouts each day, a longer mean standing bout length and drank less milk per visit. Beef-sired calves had a longer mean lying bout length and drank more slowly than their dairy sired counterparts. Intrinsic calf factors influence different healthy calf behaviors in different ways. These factors must be considered in the design of research studies and the field application of behavior-based disease detection tools in artificially reared calves.

Review: Using animal welfare to frame discussion on dairy farm technology

The use of technology on dairy farms has increased dramatically over the last half-century. The ways that scientists describe the potential benefits and risk of technology are likely to affect if it is accepted for use on farms. The aim of our study was to identify papers that describe a linkage between technologies used on dairy farms and the welfare of dairy cattle. Our systematic review identified 380 papers, of which 28 met our inclusion criteria and were used to describe the technologies examined, the welfare-relevant measures used, and the ways in which authors framed welfare benefits and risks associated with the technologies. The large majority (27 of 28 papers) used positive frames, considering how the technology could improve welfare. Some authors carefully articulated the logic linking the specific measures to specific welfare-related outcomes (such as the use of accelerometer data to draw inferences about changes in lying times), but others made more general inferences (about health and welfare) that were not (and perhaps could not) be assessed. We conclude that much of the framing focused on animal welfare is biased toward welfare benefits and that future work should strive to address both potential benefits and harms; more balanced coverage may better inform solutions to the problems encountered by the people and animals affected by the technology. Welfare is a complex and multifaced concept, and it is unlikely that any one technology (or perhaps even a combination of technologies) can adequately capture this complexity. Thus, we encourage authors to restrict their claims to specific, welfare-relevant measures that can be assessed independently and thus validated. More general claims about welfare should be treated with skepticism.

Evaluation of a Forefront Weight Scale from an Automated Calf Milk Feeder for Holstein and Crossbred Dairy and Dairy-Beef Calves

Recording of body weights of dairy calves may assist producers in monitoring the health status of calves and making feed-related management decisions. Traditional methods of weighing calves can be time-consuming and labor-intensive. The objective of this study was to evaluate a forefront weight scale on stalls attached to an automated calf milk feeder system to determine the accuracy for measuring the calf body weights of Holstein and crossbred dairy calves. The study was conducted at the University of Minnesota West Central Research and Outreach Center, Morris, MN, dairy. Eighty-eight Holstein and crossbred calves were fed either 8 L/d or ad libitum milk from September 2019 to February 2020 and March 2020 to July 2020. Crossbred calves were Grazecross crossbreds composted of Jersey, Viking Red, and Normande, ProCross crossbreds composed of Holstein, Montbéliarde, and Viking Red, Limousin-sired crossbred dairy × beef bull calves, and Limousin-sired crossbred dairy × beef heifer calves. The Limousin-sired calves were from Holstein or crossbred dams. Calves were introduced to the Holm & Laue Calf Expert and Hygiene Station automatic calf feeder (Holm & Laue GmbH & Co. KG, Westerrönfeld, Germany) at 5 days of age and were weaned at 56 d. Forefront weight scales were attached to four hygiene station feeding stalls on the automated calf milk feeder, and calves were required to place both front hooves on the scale to access milk. The calf weights from the automated milk feeder were compared to the gold standard calibrated electronic scale (Avery Weigh-Tronix LLC, Fairmont, MN scale). Calves were weighed once per week using the electronic scale, and those weights were compared to the most recent weight recorded by the forefront scale. The associations of the weights from the automated milk feeder scale and the electronic scale were determined with Pearson correlations (PROC CORR of SAS) and Bland-Altman plots (PROC SGPLOT of SAS). Furthermore, PROC GLM of SAS was used to regress the electronic scale body weight on the forefront weight scale body weight for each calf. A total of 600 weight observations were used for statistical analysis. The Pearson correlation of the electronic scale compared to the forefront weight scale was high (0.991), and the concordance correlation coefficient was high (0.987). Correlations for individual calves ranged from 0.852 to 0.999 and were classified as high. Correlations of the electronic scale and forefront weight scale for breed groups ranged from 0.990 to 0.994. The slope of the regression line was 0.9153, and the 95% confidence interval was between 0.906 and 0.925. A mean bias of 0.529 kg was observed from the Bland-Altman plots. The results suggest that there is potential for the forefront weight scale to be used on automated calf milk feeders to accurately record the body weights of calves and support management decision-making, identify sick calves, and help producers determine the proper dosage of medications for calves based on body weight.

Literature Review on Technological Applications to Monitor and Evaluate Calves' Health and Welfare

Precision livestock farming (PLF) research is rapidly increasing and has improved farmers' quality of life, animal welfare, and production efficiency. PLF research in dairy calves is still relatively recent but has grown in the last few years. Automatic milk feeding systems (AMFS) and 3D accelerometers have been the most extensively used technologies in dairy calves. However, other technologies have been emerging in dairy calves' research, such as infrared thermography (IRT), 3D cameras, ruminal bolus, and sound analysis systems, which have not been properly validated and reviewed in the scientific literature. Thus, with this review, we aimed to analyse the state-of-the-art of technological applications in calves, focusing on dairy calves. Most of the research is focused on technology to detect and predict calves' health problems and monitor pain indicators. Feeding and lying behaviours have sometimes been associated with health and welfare levels. However, a consensus opinion is still unclear since other factors, such as milk allowance, can affect these behaviours differently. Research that employed a multi-technology approach showed better results than research focusing on only a single technique. Integrating and automating different technologies with machine learning algorithms can offer more scientific knowledge and potentially help the farmers improve calves' health, performance, and welfare, if commercial applications are available, which, from the authors' knowledge, are not at the moment.

Feeding behaviors collected from automated milk feeders were associated with disease in group-housed dairy calves in the Upper Midwest United States

Automated milk feeders (AMF) are an attractive option for producers interested in adopting practices that offer greater behavioral freedom for calves and can potentially improve labor management. These feeders give farmers the opportunity to have a more flexible labor schedule and more efficiently feed group-housed calves. However, housing calves in group systems can pose challenges for monitoring calf health on an individual basis, potentially leading to increased morbidity and mortality. Feeding behavior recorded by AMF software could potentially be used as an indicator of disease. Therefore, the objective of this observational study was to investigate the association between feeding behaviors and disease in preweaning group-housed dairy calves fed with AMF. The study was conducted at a dairy farm located in the Upper Midwest United States and included a final data set of 599 Holstein heifer calves. The farm was visited on a weekly basis from May 2018, to May 2019, when calves were visually health scored and AMF data were collected. Calf health scores included calf attitude, ear position, ocular discharge, nasal discharge, hide dirtiness, cough score, and rectal temperatures. Generalized additive mixed models (GAMM) were used to identify associations between feeding behavior and disease. The final quasibinomial GAMM included the fixed (main and interactions) effects of feeding behavior at calf visit-level including milk intake (mL/d), drinking speed (mL/min), visit duration (min), rewarded (with milk being offered) and unrewarded (without milk) visits (number per day), and interval between visits (min), as well as the random effects of calf age in regard to their relationship with calf health status. Total milk intake (mL/d), drinking speed (mL/min), interval between visits (min) to the AMF, calf age (d), and rewarded visits were significantly associated with dairy calf health status. These results indicate that as total milk intake and drinking speed increased, the risk of calves being sick decreased. In contrast, as the interval between visits and age increased, the risk of calves being sick also increased. This study suggests that AMF data may be a useful screening tool for detecting disease in dairy calves. In addition, GAMM were shown to be a simple and flexible approach to modeling calf health status, as they can cope with non-normal data distribution of the response variable, capture nonlinear relationships between explanatory and response variables and accommodate random effects.

Personality of individually housed dairy-beef crossbred calves is related to performance and behavior

The aim of this study was to evaluate differences in behavioral responses of individually housed dairy-beef crossbred calves to standardized personality tests (novel person, novel object, and startle test) and investigate associations of personality traits with performance and home pen behavior. Dairy-beef crossbred (Holstein x Angus) calves (n=29) were individually housed with ad libitum access to water and calf starter. Body weight was measured weekly and calf starter intake was recorded daily from day of arrival (8.5 ± 2.1; experimental day 1) for 76 days. Behavior within the home pen (eating, drinking, non-nutritive oral manipulation) and activity were recorded on experimental days 13, 32, 53, and 67 using a camera and a pedometer. The calves were subjected to standardized personality tests in their home pen at the end of the experimental period (80.7 ± 2.0 d of age), including a novel person test (stationary person in the corner of their home pen) and combined novel object/startle test (remote-controlled car in the pen, that suddenly moved when touched). A principal component analysis on the behaviors recorded from the tests (latency to approach person or object, time spent attentive and touching the person or object, and time spent inactive, playing and grooming) yielded 3 factors that explained 76.1% of the variance, and were interpreted as personality traits labeled “fearful”, “inactive”, and “bold”. These factors were examined in regression analyses for their associations with home pen behavior and performance. The factor “fearful” had negative associations with total average daily gain and average grain intake. In contrast, the factor “inactive” had positive associations with non-nutritive oral manipulation of buckets or walls. The factor “bold” had no significant association with any of the performance or home pen behavior measures. In conclusion, personality traits identified from standardized tests were related to performance and home pen behavior measures in individually housed, crossbred calves. These results complement work in group housed calves suggesting personality testing may be useful selective tools to identify high and low performing calves from an early age.

Using Machine Learning and Behavioral Patterns Observed by Automated Feeders and Accelerometers for the Early Indication of Clinical Bovine Respiratory Disease Status in Preweaned Dairy Calves

The objective of this retrospective cohort study was to evaluate a K-nearest neighbor (KNN) algorithm to classify and indicate bovine respiratory disease (clinical BRD) status using behavioral patterns in preweaned dairy calves. Calves (N=106) were enrolled in this study, which occurred at one facility for the preweaning period. Precision dairy technologies were used to record feeding behavior with an automated feeder and activity behavior with a pedometer (automated features). Daily, calves were manually health-scored for bovine respiratory disease (clinical BRD; Wisconsin scoring system, WI, USA), and weights were taken twice weekly (manual features). All calves were also scored for ultrasonographic lung consolidation twice weekly. A clinical BRD bout (day 0) was defined as 2 scores classified as abnormal on the Wisconsin scoring system and an area of consolidated lung ≥3.0 cm2. There were 54 calves dignosed with a clinical BRD bout. Two scenarios were considered for KNN inference. In the first scenario (diagnosis scenario), the KNN algorithm classified calves as clinical BRD positive or as negative for respiratory infection. For the second scenario (preclinical BRD bout scenario), the 14 days before a clinical BRD bout was evaluated to determine if behavioral changes were indicative of calves destined for disease. Both scenarios investigated the use of automated features or manual features or both. For the diagnosis scenario, manual features had negligible improvements compared to automated features, with an accuracy of 0.95 ± 0.02 and 0.94 ± 0.02, respectively, for classifying calves as negative for respiratory infection. There was an equal accuracy of 0.98 ± 0.01 for classifying calves as sick using automated and manual features. For the preclinical BRD bout scenario, automated features were highly accurate at -6 days prior to diagnosis (0.90 ± 0.02), while manual features had low accuracy at -6 days (0.52 ± 0.03). Automated features were near perfectly accurate at -1 day before clinical BRD diagnosis compared to the high accuracy of manual features (0.86 ± 0.03). This research indicates that machine-learning algorithms accurately predict clinical BRD status at up to -6 days using a myriad of feeding behaviors and activity levels in calves. Precision dairy technologies hold the potential to indicate the BRD status in preweaned calves.

Evaluation of a point-of-care meter for measuring glucose concentrations in dairy calves: A diagnostic accuracy study

•

We validated a human glucose meter for measuring blood glucose of neonatal calves using 1,303 samples of plasma and 476 samples of whole blood.

•

For plasma, hypoglycemia was established at a threshold of <4.45 mmol/L (sensitivity, 94.2%; specificity, 91.2%).

•

For whole blood, hypoglycemia was established at a threshold of 4.95 mmol/L (sensitivity, 95.6%; specificity of 80.3%).

•

The meter can measure glycemic status in calves and may be useful for treatment decisions on farm.

The objective of this cross-sectional, diagnostic accuracy study was to validate a human blood glucose meter (Contour Next One Meter, Ascensia Diabetes Care) for accuracy and precision when measuring blood glucose, and for diagnostic accuracy for hypoglycemic status in dairy calves using whole blood and blood plasma. A total of 49 male dairy calves [body weight (BW): 46.3 ± 0.8 kg] had jugular catheters placed within 75 min after birth. Thereafter, blood was withdrawn from the catheter at specific time points (−10, 10, 20, 30, 45, 60, 90, 120, 180, 240, 360, 480, and 600 min) relative to the first and second colostrum feedings (2 h 15 min and 14 h 5 min postnatal; feeding rate: 7% of BW wt/wt). The reference standard method for plasma glucose concentration was determined colorimetrically and in duplicate using the glucose oxidase-peroxidase reaction. Data were assessed for agreement between the glucose meter and the reference standard using Lin's concordance correlation coefficients (CCC), coefficients of determination (precision), and Bland-Altman plots. In addition, a mixed linear regression model was built using the reference method as the outcome, with the glucose meter and repeated measures of time as the explanatory variables and calf as a random effect. The sensitivity (Se), specificity (Sp), and area under the curve (AUC) for the glucose meter using calf whole-blood and plasma were calculated at a threshold of <4.44 mmol/L to determine hypoglycemia. The precision (CCC = 0.95, R² = 0.93) and accuracy (AUC = 0.98) of the glucose meter were very high when used on 1,303 blood plasma samples. Youden's index revealed a threshold of <4.45 mmol/L for the glucose meter when used with plasma, leading to Se of 94.2% and Sp of 91.9%, with 92.5% of samples being correctly classified, suggesting high diagnostic accuracy. When using whole blood, precision (CCC = 0.85 and R² = 0.73) and accuracy (AUC = 0.92) were high when used on 476 samples. Youden's index revealed a threshold of <4.95 mmol/L for the glucose meter when used with whole calf blood, leading to Se of 95.6% and Sp of 80.3%, with 84.7% of samples being correctly classified, suggesting high diagnostic accuracy for use on farm. In summary, this glucose meter was validated for measuring calf blood glucose using both plasma and whole blood. This meter can measure glycemic status in calves and may be useful for clinical and on-farm use to make intervention decisions.

Social networks respond to a disease challenge in calves

Changes in network position and behavioral interactions have been linked with infectious disease in social animals. Here, we investigate the effects of an experimental disease challenge on social network centrality of group-housed Holstein bull dairy calves. Within group-housed pens (6/group) calves were randomly assigned to either a previously developed challenge model, involving inoculation with Mannheimia haemolytia (n = 12 calves; 3 calves/group) or a control involving only saline (n = 12 calves; 3 calves/group). Continuous behavioral data were recorded from video on pre-treatment baseline day and for 24 h following inoculation to describe social lying frequency and duration and all active social contact between calves. Mixed-model analysis revealed that changes in network position were related to the challenge. Compared to controls, challenged calves had reduced centrality and connectedness, baseline to challenge day. On challenge day, challenged calves were less central in the directed social contact networks (lower degree, strength and eigenvector centrality), and initiated contact (higher out-degree) with more penmates, compared to healthy calves. This finding suggests that giving rather than receiving affiliative social contact may be more beneficial for challenged calves. This is the first study demonstrating that changes in social network position coincide with an experimental challenge of a respiratory pathogen in calves.

Case-control study of behavior data from automated milk feeders in healthy or diseased dairy calves

•

Automated milk feeders have the potential to aid producers in disease detection.

•

Milk consumption detected disease 5 d before disease detection by the producer.

•

Drinking speed detected disease 4 d before disease detection by the producer.

•

Unrewarded visits to the feeder detected disease 3 d before disease detection by the producer.

•

Rewarded visits to the feeder were not useful in detection of disease by the producer.

Group housing of preweaning dairy calves is increasing in popularity throughout the dairy industry. However, it can be more difficult to individually monitor calves to identify disease in these group systems. Automated milk feeders (AMF) not only provide producers with the opportunity to increase the milk allowance offered to preweaning calves but they can also monitor individual feeding behaviors that could identify calves at increased risk of disease. The objective of this retrospective case-control study was to determine how feeding behaviors change in preweaning calves leading up to and during a disease bout. This study was conducted between fall 2015 and fall 2016 on 2 commercial dairy farms in Ontario, Canada. Producers' treatment records for respiratory or enteric illness were used to identify cases. Control calves were selected from calves not treated for disease and matched on the days on the AMF. Both farms housed calves in dynamic groups of 9 to 11 calves with an AMF and fed milk replacer. Differences in feeding behaviors, including milk consumption, drinking speed, rewarded visits, unrewarded visits, and total visits to the AMF per day, were analyzed by mixed models accounting for repeated measures. Data were analyzed for the 7 d before, the day of, and 7 d after treatment. A total of 28 cases and 28 control calves (n = 56) were analyzed. Calves with disease consumed significantly less milk than their healthy counterparts, beginning 5 d before disease and until 3 d after disease detection. Sick calves had fewer unrewarded visits starting 3 d before until 2 d after illness detection. Sick calves drank significantly more slowly starting 4 d before illness detection until the day after illness detection compared with healthy controls. No differences were found between cases and controls for rewarded visits. Calves on a high plane of milk nutrition significantly alter feeding behaviors before illness detection. Data from AMF on feeding behaviors may help to detect disease earlier in preweaning dairy calves.

Feeding behavior and activity levels are associated with recovery status in dairy calves treated with antimicrobials for Bovine Respiratory Disease

Calves with Bovine Respiratory Disease (BRD) have different feeding behavior and activity levels prior to BRD diagnosis when compared to healthy calves, but it is unknown if calves who relapse from their initial BRD diagnosis are behaviorally different from calves who recover. Using precision technologies, we aimed to identify associations of feeding behavior and activity with recovery status in dairy calves (recovered or relapsed) over the 10 days after first antimicrobial treatment for BRD. Dairy calves were health scored daily for a BRD bout (using a standard respiratory scoring system and lung ultrasonography) and received antimicrobial therapy (enrofloxacin) on day 0 of initial BRD diagnosis; 10–14 days later, recovery status was scored as either recovered or relapsed (n = 19 each). Feeding behaviors and activity were monitored using automated feeders and pedometers. Over the 10 days post-treatment, recovered calves showed improvements in starter intake and were generally more active, while relapsed calves showed sickness behaviors, including depressed feed intake, and longer lying times. These results suggest there is a new potential for precision technology devices on farms in evaluating recovery status of dairy calves that are recently treated for BRD; there is opportunity to automatically identify relapsing calves before re-emergence of clinical disease.

Validation of NEDAP Monitoring Technology for Measurements of Feeding, Rumination, Lying, and Standing Behaviors, and Comparison with Visual Observation and Video Recording in Buffaloes

The current study aimed to investigate the monitoring behaviors of the NEDAP system in buffaloes, to evaluate the validation, accuracy, and precision over visual observation and video recording. The NEDAP neck and leg tags were attached on the left side of the neck and left front leg of multiparous dairy buffaloes (n = 30). The feeding, rumination, lying, and standing behaviors were monitored by the NEDAP system, visual observation, and video recording. The feeding time monitored by NEDAP was 25.2 ± 2.7 higher (p < 0.05) than visual observation and video recording. However, the rumination, lying, and standing time was lower (p < 0.05) in buffaloes when monitored by the NEDAP technology than by visual observation and video recording. The Pearson correlation between NEDAP technology with visual observation and video recording for feeding, rumination, lying, and standing was 0.91, 0.85, 0.93, and 0.87, respectively. The concordance correlation coefficient between the NEDAP with visual observation and video recording was high for rumination and standing (0.91 for both), while moderate for feeding and lying (0.85 and 0.88, respectively). The Bland−Altman plots were created to determine the association between NEDAP and visual observation and video recording, showing no bias. Therefore, a high level of agreement was found. In conclusion, the current finding showed that the NEDAP system can be used for monitoring feeding, rumination, lying, and standing behaviors in buffaloes. Moreover, these results revealed that the buffalo behavior was monitored precisely using NEDAP technology than visual observation and video recording. This technology will be useful for the diagnosis of diseases.

Validation of an alternate method for monitoring the presence of cows at the feed bunk in a Calan Broadbent Feeding System using a 3-axis, data-logging accelerometer

Daily dry matter intake is a key observation in dairy nutrition, and observation of feeding behavior provides insight into the physiological control of hunger and satiety that regulate intake. The objective of the study was to develop and validate an alternative method to observe feeding behavior, including meal length and frequency, in a Calan Broadbent Feeding System (American Calan) using a 3-axis accelerometer (Hobo Pendant G, Onset Computer Corp.). Sensors were mounted between the door and the feed divider using commonly available materials without making permanent modifications to the feeding system. Forty-eight sensors were deployed with a recording frequency of 30 s for the last 7 d of each period in a crossover experiment with 24 multiparous and 24 primiparous animals housed in a freestall barn. The tilt angle on the Z-axis was used to determine when the door was open to indicate feeding activity. The sensor system was in very high agreement with 6 h of visual observation (Cohen’s κ = 0.92 ± 0.014; estimate ± 95% confidence interval). The minimum intermeal interval is the time between 2 feeding bouts that is still considered one meal. This essential criterion to characterize meals was calculated by determining the intersection of a mixture of Gaussian distributions fitted to the log-transformed between-feeding intervals. The best fitting mixture of Gaussian distributions was determined with the distribution module of JMP Pro 14.3.0 (SAS Institute Inc.). The minimum intermeal interval was 31.3 min using the best fitting model, a mixture of 3 Gaussian distributions. Using the determined minimum intermeal interval, meal length averaged 37.3 min/meal and meal frequency averaged 7.3 meals/d. In conclusion, data-logging 3-axis accelerometers are adequate to monitor presence of cows in the feed gate in the Calan Broadbent Feeding System, and this approach allows for reasonable estimation of meal length and frequency.

A method was developed using a 3-axis accelerometer to detect cow presence at the bunk in a Calan Broadbent feeding system to characterize feeding behavior. The sensor was at 0° when the door was closed and near 60° when the door was open, allowing clear distinction of the presence of a cow at the feed bunk, in very high agreement with visual observations. In conclusion, accelerometers provided robust monitoring of the presence of cows in the feed gate in the Calan Broadbent Feeding System, and this approach allows for economical and reasonable observation of feeding behavior.

A Systematic Review of Automatic Health Monitoring in Calves: Glimpsing the Future From Current Practice

Infectious diseases, particularly bovine respiratory disease (BRD) and neonatal calf diarrhea (NCD), are prevalent in calves. Efficient health-monitoring tools to identify such diseases on time are lacking. Common practice (i.e., health checks) often identifies sick calves at a late stage of disease or not at all. Sensor technology enables the automatic and continuous monitoring of calf physiology or behavior, potentially offering timely and precise detection of sick calves. A systematic overview of automated disease detection in calves is still lacking. The objectives of this literature review were hence: to investigate previously applied sensor validation methods used in the context of calf health, to identify sensors used on calves, the parameters these sensors monitor, and the statistical tools applied to identify diseases, to explore potential research gaps and to point to future research opportunities. To achieve these objectives, systematic literature searches were conducted. We defined four stages in the development of health-monitoring systems: (1) sensor technique, (2) data interpretation, (3) information integration, and (4) decision support. Fifty-four articles were included (stage one: 26; stage two: 19; stage three: 9; and stage four: 0). Common parameters that assess the performance of these systems are sensitivity, specificity, accuracy, precision, and negative predictive value. Gold standards that typically assess these parameters include manual measurement and manual health-assessment protocols. At stage one, automatic feeding stations, accelerometers, infrared thermography cameras, microphones, and 3-D cameras are accurate in screening behavior and physiology in calves. At stage two, changes in feeding behaviors, lying, activity, or body temperature corresponded to changes in health status, and point to health issues earlier than manual health checks. At stage three, accelerometers, thermometers, and automatic feeding stations have been integrated into one system that was shown to be able to successfully detect diseases in calves, including BRD and NCD. We discuss these findings, look into potentials at stage four, and touch upon the topic of resilience, whereby health-monitoring system might be used to detect low resilience (i.e., prone to disease but clinically healthy calves), promoting further improvements in calf health and welfare.

Lying Behaviour in Dairy Goats: Effects of a New Automated Feeding System Assessed by Accelerometer Technology

Simple Summary

Goat farming is becoming more important in Germany and as dehorning is forbidden, it is necessary to facilitate animal welfare among horned and mixed-horned herds. In this study an optimized automatic concentrated feeding system was installed in a mixed-horned herd and lying behaviour was detected by accelerometer technology. Results show a seasonal progression of lying behaviour in dairy goats and an adjustment of behavioural differences between horned and hornless goats with the new feeding system.

Abstract

The aim of this study was to evaluate lying behaviour in dairy goats before and after installation of an optimized automatic concentrated feeding system (AFS). A mixed-horned herd of Bunte Deutsche Edelziege was used. As many agonistic interactions between goats happen at the feeding place, a new automated feeding system was installed to better fulfil the needs of horned goats. Lying behaviour is an indicator to ascertain animal welfare of ruminants. In order to measure lying behaviour accelerometer technology was used and verified by video analyses. The results show an agreement of 99.62–99.93% per lying time by comparing accelerometers to video data. Over all goats, a mean ± SD lying time (LT) of 11.78 ± 1.47 h/d, a mean ± SD lying bout duration (LBD) of 0.51 ± 0.10 h/bout and a mean ± SD frequency of lying bouts (FLB) of 24.35 ± 5.57 were found. Lying behaviour follows a seasonal progression with significant lowest LBD and highest FLB in summer. With the old AFS significant differences in LBD and FLB were detected between horned and hornless goats, but with the new AFS results were adjusted. Findings suggest that changes in feeding management do not affect the general seasonal progression of lying behaviour but can affect the behavioural differences between horned and hornless dairy goats.

Dairy Cows Activity under Heat Stress: A Case Study in Spain

Heat stress plays a role in livestock production in warm climates. Heat stress conditions impair animal welfare and compromise the productive and reproductive performance of dairy cattle. Under heat stress conditions, dairy cattle modify their behavior. Thus, the assessment of behavior alterations can be an indicator of environmental or physiological anomalies. Moreover, precision livestock farming allows for the individual and constant monitoring of animal behavior, arising as a tool to assess animal welfare. The purpose of this study was to evaluate the effect of heat stress on the behavior of dairy cows using activity sensors. The study was carried out in Tinajeros (Albacete, Spain) during the summer of 2020. Activity sensors were installed in 40 cows registering 6 different behaviors. Environmental conditions (temperature and humidity) were also monitored. Hourly data was calculated for both animal behavior and environmental conditions. Temperature and Heat Index (THI) was calculated for each hour. The accumulated THI during the previous 24 h period was determined for each hour, and the hours were statistically classified in quartiles according to the accumulated THI. Two groups were defined as Q4 for no stress and Q1 for heat stress. The results showed that animal behavior was altered under heat stress conditions. Increasing THI produces an increase in general activity, changes in feeding patterns and a decrease in rumination and resting behaviors, which is detrimental to animal welfare. Daily behavioral patterns were also affected. Under heat stress conditions, a reduction in resting behavior during the warmest hours and in rumination during the night was observed. In conclusion, heat stress affected all behaviors recorded as well as the daily patterns of the cows. Precision livestock farming sensors and the modelling of daily patterns were useful tools for monitoring animal behavior and detecting changes due to heat stress.

Associations between feeding behaviors collected from an automated milk feeder and disease in group-housed dairy calves in Ontario: A cross-sectional study

The adoption of automated milk feeders and group housing of preweaning dairy calves has become more common in Canada; however, disease detection in group-housed calves remains a challenge. The aim of this cross-sectional study was to assess whether feeding behavior data collected from a single point in time could be used to aid in the detection of neonatal calf diarrhea (NCD), bovine respiratory disease (BRD), and general disease, in preweaning group-housed calves being fed via an automated milk feeder. The data used was collected in an earlier study. A total of 8 dairy farms recruited from an online survey of calf-management practices were enrolled into the study. There was a total of 523 observations with 130 events of NCD, 115 events of BRD, and 210 events of general disease. Each farm was visited once in each of the fall, winter, spring, and summer, when the calves' health was scored, and the data were collected from the automated milk feeders. Mixed linear regression models were used to identify associations between feeding behavior data (milk consumption, time spent at the feeder, drinking speed, and the number of rewarded and unrewarded visits) and the presence of NCD, BRD, or general disease (having one or more of NCD, BRD, or umbilical infection), on the day of health scoring. Generalized linear mixed models were used to analyze the percentage of milk the calf consumed from their daily milk allotment. Calves with BRD consumed 63% less of their daily allotment of milk, had 2 fewer unrewarded visits to the automated milk feeder, and drank milk 152 mL/min slower compared with calves without BRD. Calves with NCD consumed 57% less of their daily milk allotment, consumed 758 mL less per day, and drank 92 mL/min slower than calves compared with calves without NCD. Calves with general disease drank 50% less of their daily milk allowance, consumed 496 mL less per day, drank 80 mL/min slower, and had 2 fewer unrewarded visits to the automated milk feeder, when compared with calves without disease. No significant associations were found between the presence of NCD, BRD, or general disease and time spent at the feeder or number of rewarded visits. Sensitivity and specificity values for disease identification were low when evaluating the feeding behaviors individually, so parallel testing was completed. To do so, if any significant feeding behavior was below the optimal cut point for disease detection as determined using a ROC curve, the calf was considered positive for disease and the sensitivity and specificity were recalculated. Parallel testing resulted in a sensitivity of 0.82, 0.78, and 0.84, and a specificity of 0.26, 0.23, and 0.21, for BRD, NCD, and general disease, respectively. This suggests that automated milk feeders may serve as a useful preliminary tool in the detection of diseased calves. For example, producers could use feeding behavior data to identify calves requiring further inspection; however, they should not use feeding behavior data as a sole disease detection method.

Methods for Pain Assessment in Calves and Their Use for the Evaluation of Pain during Different Procedures-A Review

The evaluation and assessment of the level of pain calves are experiencing is important, as the experience of pain (e.g., due to routine husbandry procedures) severely affects the welfare of calves. Studies about the recognition of pain in calves, and especially pain management during and after common procedures, such as castration, dehorning, and disbudding, have been published. This narrative review discusses and summarizes the existing literature about methods for pain assessment in calves. First, it deals with the definition of pain and the challenges associated with the recognition of pain in calves. Then it proceeds to outline the different options and methods for subjective and objective pain assessment in calves, as described in the literature. Research data show that there are several tools suitable for the assessment of pain in calves, at least for research purposes. Finally, it concludes that for research purposes, various variables for the assessment of pain in calves are used in combination. However, there is no variable which can be used solely for the exclusive assessment of pain in calves. Also, further research is needed to describe biomarkers or variables which are easily accessible in the field practice.

Predicting morbidity and mortality using automated milk feeders: A scoping review

Automated milk feeders (AMF) are computerized systems that provide producers with a tool that can be used to more efficiently raise dairy calves and allow for easier implementation of a high plane of nutrition during the milk feeding phase. Automated milk feeders also have the ability to track individualized behavioral data, such as milk consumption, drinking speed, and the number of rewarded and unrewarded visits to the feeder, that could potentially be used to predict disease development. The objective of this scoping review was to characterize the body of literature investigating the use of AMF data to predict morbidity and mortality in dairy calves during the preweaning stage. This review lists the parameters that have been examined for associations with disease in calves and identify discrepancies found in the literature. Five databases and relevant conference proceedings were searched. Eligible studies focused on the use of behavioral parameters measured by AMF to predict morbidity or mortality in preweaned dairy calves. Two reviewers independently screened titles and abstracts from 6,675 records identified during the literature search. After title and abstract screening, 382 studies were included and then assessed at the full-text level. Of these, 56 studies fed calves using an AMF and provided some measure of morbidity or mortality. Thirteen examined AMF parameters for associations with morbidity or mortality. The studies were completed in North America (n = 6), Europe (n = 6), and New Zealand (n = 1). The studies varied in sample size, ranging from 30 to 1,052 calves with a median of 100 calves. All 13 studies included enteric disease as an outcome and 11 studies evaluated respiratory disease. Of the studies measuring enteric disease, 8 provided disease definitions (n = 8/13, 61.2%); however, for respiratory disease, only 5 provided a disease definition (n = 5/11, 45.5%). Disease definitions and thresholds varied greatly between studies, with 10 using some form of health scoring. When evaluating feeding metrics as indicators of disease, all 13 studies investigated milk consumption and 6 and 7 studies investigated drinking speed and number of rewarded and unrewarded visits, respectively. Overall, this scoping review identified that daily milk consumption, drinking speed, and rewarded and unrewarded visits may provide insight into early disease detection in preweaned dairy calves. However, the disparity in reporting of study designs and results between included studies made comparisons challenging. In addition, to aid with the interpretation of studies, standardized disease outcomes should be used to improve the utility of this primary research.

Machine Learning Algorithms to Classify and Quantify Multiple Behaviours in Dairy Calves Using a Sensor: Moving beyond Classification in Precision Livestock

Previous research has shown that sensors monitoring lying behaviours and feeding can detect early signs of ill health in calves. There is evidence to suggest that monitoring change in a single behaviour might not be enough for disease prediction. In calves, multiple behaviours such as locomotor play, self-grooming, feeding and activity whilst lying are likely to be informative. However, these behaviours can occur rarely in the real world, which means simply counting behaviours based on the prediction of a classifier can lead to overestimation. Here, we equipped thirteen pre-weaned dairy calves with collar-mounted sensors and monitored their behaviour with video cameras. Behavioural observations were recorded and merged with sensor signals. Features were calculated for 1-10-s windows and an AdaBoost ensemble learning algorithm implemented to classify behaviours. Finally, we developed an adjusted count quantification algorithm to predict the prevalence of locomotor play behaviour on a test dataset with low true prevalence (0.27%). Our algorithm identified locomotor play (99.73% accuracy), self-grooming (98.18% accuracy), ruminating (94.47% accuracy), non-nutritive suckling (94.96% accuracy), nutritive suckling (96.44% accuracy), active lying (90.38% accuracy) and non-active lying (90.38% accuracy). Our results detail recommended sampling frequencies, feature selection and window size. The quantification estimates of locomotor play behaviour were highly correlated with the true prevalence (0.97; p < 0.001) with a total overestimation of 18.97%. This study is the first to implement machine learning approaches for multi-class behaviour identification as well as behaviour quantification in calves. This has potential to contribute towards new insights to evaluate the health and welfare in calves by use of wearable sensors.