Disease early warning system to deliver big benefits

University of Nottingham School of Veterinary Medicine and Science project will use analytics and machine learning to provide a disease early warning system in calves as young as 14 weeks.
A £1.13 million project could reduce antibiotic use and save farmers million of pounds in the process.
Researchers at The University of Nottingham believe the innovation, dubbed Y-Ware, could save the UK farming industry millions of pounds, improve animal health and welfare and reduce antimicrobial use.
Bolus sensors
To date, bolus sensors, which sit in the animal’s gut and monitor body temperature or pH, are the most widespread piece of kit used, but are only available for adult cows.
The Y-Ware project aims to develop a bolus sensor that could be used in calves as young as 14 weeks, in addition to a dashboard to collate information from a range of sources, including temperature and veterinary records, to provide farmers with a comprehensive early warning system.
Data signatures
This will be used to produce baseline data and a specific “signature” for each animal. Any changes, such as an unexpected rise in body temperature, could allow farmers to spot signs of disease, and treat it, early. In turn, farmers will be able to more effectively target use of antibiotics to treat such diseases.
The consortium is made up of specialists in engineering technology, software development, veterinary epidemiology, cattle health and data science, cloud computing and data analytics.
The project is a partnership with farming digitalisation specialists PrognostiX and BT, and is supported by a grant from Innovate UK, the UK Government-funded innovation agency.
Innovative technology
Jasmeet Kaler, associate professor of epidemiology and farm animal health in The University of Nottingham School of Veterinary Medicine and Science, is the academic lead on the project.
Dr Kaler said: “In this project, we are leading data analytics working alongside our partners. We will use our domain knowledge with regard to our understanding of disease biology and epidemiology, together with various machine learning approaches on the data gathered via sensors. Our overall aim will be to develop an innovative technology that combines different formats of data, uses application of the ‘internet of things’ and advanced analytics for early detection of disease in young stock, and, thus, allow targeted use of antibiotics.”