Facts:
Mohammed Sallam defends his thesis on the 7:th of March. Link to the doctoral thesis Digital and genetic tools to improve bones of laying hens | SLU publication database (SLUpub)
News
Digital and Genetic Tools Can Improve the Breastbone Health of Laying Hens
Published: 05 March 2025
Up to 90 percent of the world’s laying hens suffer from breastbone injuries—painful fractures and deformities that affect both animal welfare and egg production. Now, researchers at the Swedish University of Agricultural Sciences (SLU) have developed an AI-based method that can efficiently identify and analyze breastbone quality. This method could pave the way for healthier hens and a more sustainable egg industry.
"Our AI-based method enables efficient and objective our AI-based method enables efficient and objective measurements, based on the shape of the breastbone, that are very good predictors for the presence of injuries. This opens up new possibilities for selecting laying hens with improved bone health, which enhances both animal welfare and the sustainability of egg production," says Sallam Mohammad Abdallah, PhD student at SLU.
Globally, there are approximately 7.8 billion laying hens producing 87 million tons of eggs annually. In Sweden alone, 7.7 million hens contribute to 118,000 tons of eggs. Despite the scale of production, a significant proportion of hens suffer from breastbone injuries—an issue that has been difficult to prevent.
Traditionally, these injuries have been linked to osteoporosis, but research shows that even hens with strong bones can develop fractures. This suggests that additional factors, such as the shape and structure of the breastbone, play a crucial role. To better understand the causes of these injuries, SLU researchers have developed a method that combines AI and X-ray imaging to analyze breastbone quality.
Using deep learning and computer vision, the AI-based method can assess breastbone shape and structure from X-ray images. In a study of 1,050 birds, researchers examined different measures of breastbone quality. The results showed that bone density was not a reliable indicator of injury, likely because fractures can lead to the formation of callus tissue around the damaged area. However, the concave area of the breastbone—a measure of its curvature—proved to be a relevant factor. A larger concave area was associated with more fractures and deformities. This trait also exhibited moderate heritability and a genetic correlation with fracture occurrence, making it a promising characteristic for selection in breeding programs.
“The new method can also be developed to analyze other bones, such as the tibia, and has potential applications in both breeding programs and research on nutrition and housing conditions. By combining digital technology with genetic selection, we can improve animal welfare and create a more sustainable egg production system in the future,” says Sallam Mohammad Abdallah.
Sallam Mohammad Abdallah, photo: Erika Troeng