The heart of a performance horse is an extraordinary but delicate engine. While many irregular rhythms are harmless, severe cases can lead to sudden performance failure, trackside collapse, or even fatal cardiac episodes. Until now, identifying these dangers usually required monitoring the animal during peak exertion—a process that is both logistically difficult and potentially risky for a horse already at at-risk. This new method shifts the focus to prevention, analyzing the "disorderliness" of the heart’s electrical signals while the horse is either at rest or engaged in light exercise, with a heart rate between 60 and 100 beats per minute.

At the core of this discovery is an artificial intelligence system developed by Surrey researchers in collaboration with the University of Minnesota and leading veterinary clinics. By analyzing 60-second ECG segments from a group of 110 Thoroughbreds and Standardbreds, the AI demonstrated remarkable predictive power, achieving an accuracy score (AUC) of 0.86. This high level of precision means the system is not only effective at spotting vulnerable horses but is also highly reliable at ruling out healthy ones, preventing unnecessary medical interventions.

According to Professor Kamalan Jeevaratnam, Head of the School of Veterinary Medicine at Surrey, this approach fundamentally changes how we handle animal welfare. Rather than waiting for a crisis to manifest during a high-stakes race, veterinarians can now pick up warning signs during a standard warm-up, allowing for much closer monitoring of vulnerable athletes.

This technology does not exist in a vacuum; it builds upon the team's previous research into paroxysmal atrial fibrillation, the most common form of sustained arrhythmia in both horses and humans. By combining these techniques, vets may soon have a comprehensive diagnostic tool capable of screening for multiple life-threatening conditions in a single, low-intensity session, ensuring a safer future for the horses and the entire racing industry.