Artificial intelligence could predict bone strength in chickens
The tool could identify birds or family lines with stronger bones for egg-laying.
Researchers have explored how artificial intelligence (AI) could be used to inform breeding decisions for chickens’ bone health.
A team from the Roslin Institute has created an automated tool which is able to analyse X-ray images to estimate the strength of leg bones, which are an overall indicator of skeletal health.
The tool was created as researchers sought to assess diagnosis of bone damage, particularly in laying hens. These birds can develop fractures as a result of the physical demands of egg production and movement in housing systems.
Its AI model has been trained to identify a chicken’s tibia bone from an X-ray image, and then analyse patterns to predict how strong the bone might be. It was developed using more than 900 X-ray images.
Results revealed a strong correlation with bone strength measured using standard manual tests. The tool also outperformed the manual scoring of X-rays, raising possibilities for reducing time and labour involved.
Since it uses image data alone, and predictions would be generated quickly, the AI tool could enable the large-scale assessment of bone strength.
While the tool has been trained using post-mortem samples, the team hopes to apply the method to live birds in future. With measurements closely matching the underlying genetics of bone strength, it is expected that the tool could identify birds or family lines with stronger bones.
This will mean that, although the poultry breeding industry currently has to assess large numbers of birds to guide breeding decisions, the tool could provide a rapid, non-invasive and scalable alternative.
The tool has been based on widely used AI systems, which have been adapted for the task. This means that it can be implemented into standard computing infrastructure and be accessible for use in breeding programmes.
Tanmay Debnath, PhD student and lead author, said: “We’ve shown that we can use X-ray images to generate a measure of bone strength that is just as useful as conventional methods.
“From a genetic perspective, the results are very strong, birds identified by the AI as having stronger bones strongly overlap those we would select using traditional testing.”
The full study has been published in the journal Poultry Science.
Image © sergey kolesnikov/Shutterstock.com
Eligible members of BEVA will be able to vote for their Council team until Monday, 17 August 2026. Members will have received an election email on 17 July.
