Researchers from the University of Bristol (UoB) have received funding to investigate whether artificial intelligence (AI) could be used to detect disease in dairy cows earlier.
The project will see researchers use AI to monitor social interactions of cattle that could indicate developing mastitis or lameness and enable the diseases to be caught at an earlier stage.
UoB researchers will receive a share of £9 million in funding from the Biotechnology and Biological Sciences Research Council (BBSRC) and the Department for Environment, Food and Rural Affairs (Defra) as part of a joint-initiative to combat endemic livestock disease.
The UoB project is a collaboration of researchers in veterinary medicine, animal behaviour, computer vision and AI, led by Prof. Andrew Dowsey from Bristol Veterinary School.
While technologies do exist that detect disease in dairy cows, these tend to focus on observable symptoms which are associated with later stages of disease, so Dowsey and his colleagues are seeking to utilise AI to identify diseases at an earlier stage.
“A cow’s response to infection or trauma is to reduce behaviours which are not immediately essential to survival, such as social interactions,” Dowsey said.
“In a recent study we found that social exploration, the grooming of others, and receiving headbutts were all lower in cows with early stage mastitis, so we think social behaviour changes could be early predictors of disease.”
Researchers said they know detecting social behaviour changes is difficult for a busy farmer, but say that it is possible when monitoring them at key points such as queuing for milking or feeding time.
Dowsey and his team have developed an AI that can track the motion of cows, recognising each cow by its distinctive coat pattern.
“From collecting two years of video from 64 cameras covering our main barn at the John Oldacre Centre dairy farm, we will train a model that learns what types of behaviours change over time that are indicative of early-stage mastitis and lameness,” he said.
The system will be then be deployed at a network of recruited farms for testing.
AI developments
Dowsey said he was delighted that the UoB research team’s funding bid has been successful as they seek to further develop the potential uses for AI.
“We are excited to have been given the opportunity to fuse Bristol Veterinary School’s world-class expertise in animal behaviour with our recent developments in artificial intelligence for livestock monitoring,” he said.
“Detecting subtle changes in social behaviour could hold the key to the early diagnosis of disease in dairy cattle.”
The UoB project is part of a wider BBSRC and Defra initiative aiming to lessen the burden of endemic diseases on animal health, welfare, and productivity across the whole UK livestock sector.
The endemic livestock disease initiative focuses on fostering collaborative research between academia, industry and farmers and veterinarians.
The projects chosen for funding cover all areas of the livestock sector – covering pigs, poultry, beef, sheep and dairy.
Interim executive char at BBSRC, Prof. Guy Poppy, said: “Endemic diseases in the UK livestock sector pose significant challenges to animal welfare, productivity and sustainable farming practices.
“By bringing together the collective expertise of academia, industry and end-users, we are confident this initiative will lead to ground-breaking advancements in disease control, fostering a healthier and more productive livestock sector.”