Surrey AI flags racehorse heart risks from short ECG
Get the Health newsletter
Daily health & science — research, biotech, public health, the studies worth knowing. Free.
- University of Surrey researchers found that a short ECG taken during a warm-up trot can flag racehorses at risk of exercise-induced cardiac arrhythmias, potentially preventing cardiac events in otherwise healthy horses showing no obvious warning signs.
- The team developed an AI system that measures 'disorderliness' of the heart's electrical signal and can distinguish horses likely to develop premature beats at peak exercise using only short ECG recordings at heart rates of 60–100 bpm.
- Published in Scientific Reports, the study recorded ambulatory ECGs from 110 US-based Thoroughbred and Standardbred racehorses during routine training, with the best-performing algorithm achieving an area under the curve of 0.86—well above the 0.5 chance threshold.
- The test was highly effective at ruling out horses not at risk, meaning positive screens could be referred for a full exercising ECG to confirm the type and severity of any underlying arrhythmia.
- Professor Kamalan Jeevaratnam, Head of School of Veterinary Medicine at the University of Surrey, said the method lets vets flag at-risk horses during routine warm-up rather than waiting for an arrhythmia to manifest at peak intensity.
- The study builds on the Jeevaratnam group's previous work detecting paroxysmal atrial fibrillation; the two methods operate at different heart rate ranges, potentially allowing vets to screen for both conditions within a single low-intensity diagnostic session.
Why it matters: The 0.86 AUC on 110 horses establishes a practical pre-race screening tool for an industry where exercise-induced arrhythmias can cause collapse or sudden cardiac episodes. Vets and trainers gain a 60-second warm-up check that flags at-risk horses before peak exertion, with positive screens referred for full exercising ECG confirmation under the study's proposed workflow.
Ask SkimNews
