Newly born piglets face a particular challenge—the sow that births them often does not have as many teats as she has piglets. In the US, up to one in four piglets die before they have completed nursing. This high preweaning mortality is due to multiple factors that bring economic challenges for producers, take an emotional toll on caretakers, and represent an animal welfare issue.
Developing best practices for the industry is complicated by the multiple factors contributing to preweaning piglet mortality, which include difficulties associated with labor needs and staff training, the complex nature of sow-piglet interactions, and the piglets’ instinct to huddle on one another.
Dr. Madonna Benjamin, with MSU co-investigators Dr. Daniel Morris and Elizabeth Ferry, and an international team, are using artificial intelligence to better understand the feeding patterns of piglets and, ultimately, develop solutions to decrease preweaning mortality, improve production efficiency, and ensure the welfare of all piglets.
The team’s research project, IDEAS Tripartite: Automated Piglet and Sow Monitoring for Early Detection of At-Risk Piglets, has been awarded a four-year, million-dollar grant by the National Institute of Food and Agriculture (NIFA). The funding comes through NIFA’s Inter-Disciplinary Engagement in Animal Systems (IDEAS) program, which aims to address the “complexity of social, cultural, environmental, economic, and technologic challenges facing the food and agriculture system” by supporting projects that “bridge traditional disciplinary divides.”
Benjamin recruited top researchers from around the globe for the multidisciplinary team. The project team comprises scientists and computer vision engineers from North America, the Republic of Ireland, and Northern Ireland.
“Our goal is to capture a ground truth on piglet nursing that will inform not only our own research but will be available for researchers of all disciplines to use,” says Benjamin. “Through
the engineering expertise of team members, the computer vision capture systems will see nuances that human vision cannot—each camera focusing on just one sow and her piglets for 24 hours a day without the distraction of other duties.”
Benjamin expects this baseline data to provide lots of opportunities for researchers from fields including nutrition science, breeding, lactation biology, and animal sciences such as genetics and genomics.
Earlier research by team members has shown that piglet survival and growth are dependent on access to a functional teat for colostrum and milk. Producers and stockpersons tend to identify the number of teats when the sow gives birth. Yet, non-functional teats may not be visually evident until day three of lactation. If the team can quantify teat function, and if this information can help pinpoint non-functional teats, it would be an early identifier that could alert the producer to the exact number of available teats and whether there are enough teats for piglets.
Productivity, Animal Husbandry, Data
Across farms in the US, there is currently little consistency in standard operating procedures, which reduces the chances of any consistent improvement in piglet vitality. Data is required to inspire any large-scale shift in practices.
Knowledge generated by Benjamin and her team is intended to provide solid data that can inform sow and piglet husbandry practices—in North America and beyond. They are recruiting producers and stakeholders for their insights, which will be important for the ultimate adoption of any best practices determined by this research. The team will also survey caretaker opinions of the use of artificial intelligence in job satisfaction. The researchers call it “Willingness to Play.” As a grand finale, the team will publicly showcase findings in an exhibit called PigLetArt. The show, to be hosted on four international campuses, will include large-scale photos and videos from the research, coordinated with interactive educational demonstrations.
“NIFA–IDEAS reviewers described our proposal as ambitious,” says Benjamin.
Indeed, the unique constellation of experts achieved a proposal reflecting the aim of the IDEAS program —a creative and innovative solution-focused project that promises to provide solutions to complex issues faced in animal agriculture.
This work is supported by the Food Research Initiative’s Inter-Disciplinary Engagement in Animal Systems (A1261) from the U.S. Department of Agriculture’s National Institute of Food and Agriculture.
