AI Monitoring System for Calves’ Health
UNIVERSITY PARK, Pa. — Bovine respiratory disease (BRD), a serious type of pneumonia, poses a significant threat to dairy calves, leading to substantial economic losses exceeding $1 billion annually for the U.S. cattle industry. To combat this issue, a team of researchers from Penn State, the University of Kentucky, and the University of Delaware has received a three-year, $1 million grant from the U.S. National Science Foundation to develop a cutting-edge monitoring system that leverages advanced artificial intelligence (AI) and modern sensing technologies. The system aims to detect BRD in its early stages, potentially saving the lives of calves and reducing reliance on antibiotics.
Melissa Cantor, assistant professor of precision dairy science and lead collaborator at Penn State’s College of Agricultural Sciences, emphasizes the importance of early detection in preventing illness. “We know that early detection can save the lives of calves, reduce antibiotic use, and improve farmers’ profitability,” she stated. The innovative system, named CalfHealth, will incorporate wearable sensors and robotic smart feeders to monitor calf health proactively. Additionally, the project will focus on strategies to encourage farmer adoption and trust in this integrative system.
The team, which includes co-principal investigators Simone Silvestri from the University of Kentucky and Michelle Segovia from the University of Delaware, is building on previous research into calf health monitoring. According to Cantor, CalfHealth will utilize multimodal detection methods, meaning it will collect various types of data through low-cost sensors that calves will wear, including accelerometers to monitor physical activity and resting behavior. Furthermore, precision robotic feeders will help track feeding patterns, and a non-invasive Wi-Fi-based sensing system will analyze breathing patterns.
To enhance detection accuracy, the AI system will employ a deep-learning technique known as attention mechanisms, focusing on significant behavioral or breathing changes indicative of illness. Cantor highlights the flexibility of the system, indicating it will adapt to various farming conditions while maintaining effectiveness — a crucial aspect for real-world implementation.
Beyond benefiting dairy farming, the developments promised by CalfHealth could extend to beef cattle and support the early identification of other diseases, such as diarrhea or even avian influenza outbreaks, provided that farmers can effectively integrate these tools. To facilitate understanding, the system will include an interactive chatbot powered by advanced language models to explain the rationale behind health alerts and answer farmers’ questions regarding their livestock’s health.
Research on the interaction between farmers and AI tools will be incorporated into the project, examining how trust can be cultivated through behavioral science. Plans also include testing CalfHealth on various farms and evaluating the impact of early detection on calf health and farm profitability. Workshops and community demonstrations will be organized to engage farmers, veterinarians, and stakeholders to promote awareness and understanding of the technology.
An investment in this type of innovative research highlights the critical challenges facing agricultural industries today, as federal support has historically fueled such technological advancements. However, ongoing federal funding cuts pose a significant risk to the future of research and development in agriculture and other vital sectors.
Learn more about the implications of federal funding cuts to our future at Research or Regress.