Researchers develop AI system that can detect fractures in racehorses.

Research by the Royal Veterinary College has identified that Artificial Intelligence (AI) has the potential to transform fracture detection in animals. 

The study, published in the journal Bioengineering, revealed that AI can detect and localise fractures in horses by employing knowledge from thousands of images of human fractures.

Led by Dr Ruby Chang, associate professor of statistics at the RVC, the approach is known as ‘transfer learning’, and could pave the way for AI-assisted tools to strengthen fracture diagnosis across veterinary practice.

For the study, Dr Hanya Ahmed compiled a bank of images comprising 100 equine fracture cases from two UK equine hospitals and published literature; 70 feline cases from hospital databases; and approximately 4,000 human fracture images from a public database.

Using these images, Dr Ahmed created an AI system that works in three stages: first, identifying the type of scan (X-ray, CT or MRI), then recognising the image angle, before detecting and precisely locating any fractures.

Dr Ahmed’s approach enabled the model to be trained on a large human dataset before being adapted for veterinary use. As a result, the system achieved fracture localisation accuracy ranging from 71 and 84 per cent, without requiring an excessively large number of equine images.

It is hoped that the findings could lead to faster and more reliable detection of fractures. This could help reduce uncertainty in clinical decision-making and enable earlier treatment, with clear benefits for the welfare and recovery of racehorses and companion animals.

In recognition of the work, the study has been shortlisted for the prestigious STEM for Britain 2026 award.

Dr Chang said: “I am delighted that research from our team, led by the outstanding work of Dr Hanya Ahmed, has been selected as a finalist for the prestigious STEM for Britain 2026. Dr Ahmed has brilliantly translated expertise in medical image analysis to the veterinary field, developing a novel AI system to detect fractures in racehorses.

“This exceptional work has now also been published in Bioengineering. This dual recognition is a testament to Dr Ahmed's skill and dedication, and a wonderful celebration of our team's collaborative effort to advance diagnostic technology.”

Image (C) Shutterstock/Lukas Gojda.