Recipient Organization
UNIVERSITY OF KENTUCKY
500 S LIMESTONE 109 KINKEAD HALL
LEXINGTON,KY 40526-0001
Performing Department
Animal and Food Sciences
Non Technical Summary
Good animal welfare for farm animalsis important to all in the dairy industry, including producers, processors, distributors, and cooperatives. The development of a way to automatically evaluatethe animalwelfare level of the cattle on-farm is fundamental. Technologies that monitor the physiology, behavior and perfomance of the animals at the individual animal level already exist. Before these technologies can be useful in assessing animal welfare, predictive models and validations must first be done. Beyond the monitoring of animal welfare level on farms, the public has to trust the system and be engaged in the process. Therefore, the public must be engaged to establish which aspects of these technologies may generate social acceptance or concern. Thus, our proposed integrated research and extension project aims to bridge the use of technologies that monitor each animal on a dairy farm with the social aspects of animal welfare. We will develop models and validate the use of multiple, integrated technologies to predict common animal welfare assessment outcomes that can be monitored remotely while simultaneously engaging dairy producers and the public in two-way conversations about the role of these technologies on-farm. Our multidisciplinary project will integrate scientific assessments of animal welfare, artificial intelligence, machine learning, dairy production knowledge, and social science to provide practical recommendations for the sustainableuse of technologies that monitor each animal on-farm.
Animal Health Component
50%
Research Effort Categories
Basic
50%
Applied
50%
Developmental
0%
Goals / Objectives
Good animal welfare is paramount to the dairy industry, including producers, processors, distributors, and cooperatives. The development of a new, accurate, and remote welfare assessment benchmark using validated multi-variable precision dairy technologies (PDTs) has the potential to increase the sustainability of the dairy industry. PDTs allow for real-time, continuous recording of animal behavior and other animal-based outcomes at the individual animal level. Before these technologies can be useful in assessing animal welfare, predictive models and validations must first be done. Additionally, although technology may be useful to identify animal welfare concerns on-farm, dairy producers must be willing to adopt these technologies, see value and trust in these tools, and interpret the data. Concurrently, there is a risk that investment in and adoption of novel technologies may be futile if these technologies are ultimately rejected by society. Therefore, the public must be engaged to establish which aspects of these technologies may generate social acceptance or concern. Thus, our proposed integrated research and extension project aims to bridge the use of PDTs with the social aspects of animal welfare. We will develop models and validate the use of multiple, integrated technologies to predict common animal welfare assessment outcomes that can be monitored remotely while simultaneously engaging dairy producers and the public in two-way conversations about the role of these technologies on-farm. Our multidisciplinary project will integrate scientific assessments of animal welfare, artificial intelligence, machine learning, dairy production knowledge, and social science to provide practical recommendations for the sustainable use of PDT on-farm.Long-term Goal and Supporting ObjectivesOur long-term goal is to create evidence-based recommendations for the sustainable use of precision technologies on dairy farms, with a focus on improving animal welfare assessment and individual animal management on-farm using PDT data. We intend to accomplish this goal by pursuing 3 specific objectives:? Objective 1: Develop and validate the use of PDTs to predict animal-based measurements collected manually from animal welfare assessments (e.g., lameness, injuries, and body condition) using algorithms created from machine learning. We predict that the PDTs, in combination with variables routinely collected from farms, will yield precise and accurate data comparable to manual assessments.? Objective 2: Identify key variables collected by PDTs that best predict animal-based outcomes at the herd- and cow-level. We predict that a combination of variables from the PDTs will be able to accurately detect welfare concerns at the herd- and cow-level.? Objective 3: Identify support, concerns, and trust relative to PDTs among producers, other industry stakeholders (e.g. processors, veterinarians, nutritionists, consultants, and extension specialists) and the public, as well as disseminate project findings to dairy producers using extension efforts. We predict that increased familiarity and use of data collected by PDTs will improve producer and other industry stakeholder trust in PDTs as a component of management and animal welfare assessments. Public acceptance is predicted to associate with the extent to which PDTs are perceived to confer welfare improvements to animals.
Project Methods
MethodsObjective 1: Develop and validate the use of PDTs to predict animal-based measurements collected manually from animal welfare assessments (e.g., lameness, injuries, and body condition) using algorithms created from machine learning.Facilities and Recruitment. To reach this objective,dairy farms distributed in the Southeast and Midwest of the U.S. will be enrolled in one study. Recruitment will be performed by our team.Animals and Study Design. A subset of 40 dairy cows from each farm will be selected following criteria based on Van Os et al. (2019). Cows will be enrolled at least 20 days before expected calving dates, and data will be collected from d 0 to 305 after calving. Two validated PDTs (a collar on each cow and a 3D camera installed on a walkway exiting the milking parlor) will be provided to each farm. The collar will measure activity, rumination, and the time devoted to resting (time stopped without ruminating; Grinter at al., 2019), and the 3D camera will measure body condition scores (1-5 scale, 0.25 BCS increments; Mullins et al., 2019). Data will be collected from the PDTs twice per day for BCS (at milking), and hourly for activity, rest, and rumination time. The research group, with the aid of service representatives of the corresponding technology, will be available for troubleshooting or replacement of failed technology throughout the project.In addition to data collected from the PDTs, the research team will also collect other data routinely collected on-farm from each of the 40 cows (milk yield, milk components, and health records). Milk yield and health records will be collected using the farm's management program (e.g., DairyComp305 and PCDart), and milk components will be measured by taking a milk sample from each cow and evaluating the components in-house during the evaluation visits.During the first farm visit and every 4 months thereafter until 305 d (months 0, 4, and 8), two trained evaluators from the study team will conduct an animal care assessment adopted from the National Dairy F.A.R.M Program ("Manual Assessment"). The entire lactating herd, including hospital and special needs pens, will be scored. Each farm will be categorized as "compliant" or "needs improvement" using the F.A.R.M. program's thresholds for continuous improvement. If a farm falls below the threshold for one of these animal-based measures, they will be categorized as "needs improvement".Model Development. Thresholds for animal-based outcomes used by the F.A.R.M program will be applied as acceptance thresholds indicating farms that require improvement to their management ("needs improvement"). Using these thresholds as the 'gold standard', a machine learning model will be created using the behavioral data collected from the collars, the BCS data collected from the camera, as well as other routinely collected data from the farm (milk yield, milk components and health records). A subset of data will serve as inputs into a machine learning algorithm using artificial intelligence networking to create a model that will assess the accuracy of the PDTs and routinely collected data in comparison to the Manual Assessments.Objective 2: Identify key variables collected by PDTs that best predict animal-based outcomes at the herd- and cow-level.Animals and Study Design. The second objective of this study is to streamline and optimize the machine learning model created in Objective 1 to develop a simple, efficient, and accurate model to predict herd- and cow-level welfare concerns. The same animals, farms, and data used in Objective 1 will be used to reach this objective.Design and Model Development. To build off of Objective 1, we will develop a model to determine the accuracy of our PDTs at detecting animal welfare concerns at the cow level (e.g., lameness scores of 3, hock/knee lesion scores of 3, a broken tail, and BCS scores <2) measured using the Manual Assessment. Machine learning will be used to combine the results of the PDTs and routinely collected farm data into data clusters that best predict animals with a welfare concern (e.g., lameness scores of 3, hock/knee lesion scores of 3, a broken tail, and BCS scores <2). These data clusters will be adjusted and streamlined to only contain the variables that most effectively will be able to accurately predict at-risk animals based on previously known on-farm welfare assessment data.Objective 3: Identify support, concerns, and trust relative to PDTs among producers, other industry stakeholders, and the public, as well as disseminate project findings to dairy producers using Extension efforts.?Identify producer perspectives toward and willingness to trust PDT as a component of animal welfare assessments. To reach this aim we will conduct 2 producer surveys, one using the 50 producers already recruited for Objective 1, and one using a larger sample of producers across the U.S., as well as focus groups. Producers applying PDTs on their farms during their involvement in Objective 1 will be the target population for a researcher-administered survey designed to capture changes in producers' perspectives of PDTs over time. The survey will be administered at three intervals: a) 1 month before the technology is installed on the dairy to determine producers' initial perception of the role and ability of PDTs as welfare assessment tools; b) after 4 months of PDT use; and c) after 8 months of PDT use. Survey variables will facilitate comparison of producers' expectations of technology before and after engagement with it, and how their perceptions of animal welfare assessment are affected by the use of the technology.Quantitative findings from the survey will be explored in more detail using six focus groups, three each in KY and TN. The focus groups will be used to explore factors that contribute to changes in perception of PDTs over time and determine producer perspectives on the role of technology in animal welfare assessment. The goal of the focus groups is to better understand producers' perceptions of the utility of the PDTs and how the PDTs could affect their comfort with animal welfare assessments in the future. Focus groups will occur in conjunction with existing producer meetings to facilitate producer travel.?Explore public acceptance of and concerns about the use of PDTs on-farm. A national online survey will be deployed in the second half of year 1 to determine: 1) public acceptance of and concerns relative to the use of PDTs to automate welfare assessment on dairy farms and 2) the effect of framing of information about PDTs (e.g. as benefiting cow welfare, ease of producer management, or economic production costs) on public acceptance of these technologies. The survey will be hosted on Qualtrics and participants will be recruited through the Amazon Mechanical Turk crowdsourcing platform.?Engage southeastern dairy producers in the potential contributions of PDTs to manage dairy animals and to augment animal welfare assessments. Through extensive contacts with dairy producers, Extension workers, cooperatives, and veterinarians, we expect to provide opportunities for producers to discuss and demonstrate promising strategies to utilize PDT to improve the management of dairy animals on-farm. Beyond the extension outreach material and other activities planned in this project, to further engage dairy producers locally, we plan to create four "demonstration" farms (two participating producers as well as the University of Kentucky and University of Tennessee dairy farms) to host at least 2 "field days" each for interested producers and stakeholders in the region and beyond. Producer questions and concerns identified during the field days also will be recorded and then addressed in the next events and all outreach materials.?