A research project combining artificial intelligence and animal welfare
Duration: 9m 29s
Liette Lamonde, General Manager of Prompt talks to Élise Gosselin from Novalait and Abdoulaye Diallo of UQAM about a research project focusing on animal welfare.
In recent years, dairy farmers have been working hard to improve their productivity, particularly in terms of herd feeding. The research project that Prompt is presenting today goes a step further, integrating artificial intelligence to improve animal welfare. Because animal welfare also means productivity.
Here are a few extracts:
(Liette) The aim of your project is to evaluate how changes in the physical environment, handling and management of cows influence their well-being and longevity. What was the need behind this research project? What was the problem we were trying to solve, or are still trying to solve because the project is ongoing?
At Novalait, animal welfare has been a priority for over 25 years. It’s crucial for producers to ensure that they apply best practices on the farm. Our research teams have to manage a massive amount of data, including videos, to study cow behavior. This task was becoming cumbersome and problematic. Commercial farms also use cameras to monitor the animals.
Abdoulaye, from your point of view as a researcher, what’s the problem you’re trying to solve?
We collect huge amounts of on-farm data to analyze cow behavior. A major challenge is to identify welfare indicators from the videos, and to deploy systems capable of capturing these indicators in real time. This involves analyzing videos, extracting signals, managing security and privacy issues, and using AI tools to extract properties related to animal welfare. So we use technology to improve animal welfare.
Do you have a concrete example of something we’d like to observe that could help the cow’s well-being?
Yes, in particular, we’re looking for early signs of, perhaps, problems with cow mobility or problems with cow feet and limbs. Cows, by their very nature, are prey, so they tend not to express much discomfort or pain, which means that when we see a cow limping a little, it may already be a little advanced.
What lies behind this collaboration? Because there are several players involved.
[…] we have people in animal science who will have an understanding of artificial intelligence, which is an asset, and we have people in artificial intelligence who are happy to put both feet on the cow floor and see that what they can do in a way that is sometimes a little theoretical or, but it can have a resonance in the daily lives of animals and then dairy producers to improve welfare. So that’s really interesting. And indeed, we found this idea very good.
What motivated you to innovate in this project, Abdoulaye?
We’re motivated by the pleasure of seeing our technologies make a difference, for example by reducing the time students have to spend working, from 20-30 hours to 5 minutes measuring an indicator. This improves the quality of research. Producers also see a promising future with AI and are becoming less reluctant to use these technologies in their sector.
Project partners: Novalait,UQAM,McGill University, Dairy Farmers of Canada, Dairy Farmers of Ontario and Lactanet.
