Bridging Species with AI: A Cross-Species Deep Learning Model for Fracture Detection and Beyond
Hanya T Ahmed1, Dagmar Berner2, Qianni Zhang3
1Department of Comparative Biomedical Sciences, Royal Veterinary College, London NW1 0TU, UK.
Bioengineering (Basel, Switzerland)
|February 27, 2026
Summary
A new deep learning model aids fracture detection in racehorses by using transfer learning from human data. This AI approach improves diagnostic accuracy and can be adapted for various veterinary and human health applications.
Area of Science:
- Veterinary Medicine
- Artificial Intelligence
- Medical Imaging
Background:
- Fractures are a major cause of illness and death in Thoroughbred racehorses, impacting their welfare and athletic careers.
- Current diagnostic methods for equine fractures can be time-consuming and require specialized expertise.
Purpose of the Study:
- To develop and validate a deep learning model for accurate fracture detection in equine athletes.
- To explore the efficacy of transfer learning from human medical data for veterinary applications.
Main Methods:
- A deep learning architecture combining a Vision Transformer and ResNet backbone was utilized.
- The model was trained on a diverse dataset of equine radiographs, including fracture and non-fracture cases.
- Transfer learning principles were applied, leveraging prior training on human fracture data.
Main Results:
- The model achieved high accuracy in modality classification (96.7%) and projection recognition (97.2%).
- Fracture localization performance demonstrated intersection over union values ranging from 0.71 to 0.84.
- The study confirmed the effectiveness of cross-species transfer learning for medical image analysis.
Conclusions:
- The developed deep learning model shows significant promise for improving the precision and efficiency of equine fracture diagnosis.
- This AI-driven approach has the potential for broader applications in veterinary diagnostics and human healthcare.
- The study establishes a foundation for versatile, AI-powered health monitoring systems across different species.
Related Concept Videos
01:27Fractures: Bone Repair
5.9K
Treatment for a fracture is based on the type of break, the bone affected, and the patient's age.
Minor fractures with no bone displacement are treated by immobilizing the fractured bone using a cast or splint. However, in the case of fractures with displaced bones, the broken bones are repositioned before immobilization to ensure successful healing without deformation and loss of function. The realignment of fractured bone ends is performed through a process called reduction. If the...
Minor fractures with no bone displacement are treated by immobilizing the fractured bone using a cast or splint. However, in the case of fractures with displaced bones, the broken bones are repositioned before immobilization to ensure successful healing without deformation and loss of function. The realignment of fractured bone ends is performed through a process called reduction. If the...
5.9K
01:18Classification of Bones
10.5K
The bones of the human skeletal system are of varied shapes, sizes, and functions. They can be classified based on their shape and function into four major classes: long bones, short bones, flat bones, and irregular bones. Some classifications include a fifth type, the sesamoid bones, as a separate class, whereas others categorize them under short bones.
Long and Short Bones
The appendicular skeleton, particularly the upper and lower limbs, is primarily made of long and short bones. The...
Long and Short Bones
The appendicular skeleton, particularly the upper and lower limbs, is primarily made of long and short bones. The...
10.5K