Artificial Intelligence-Enhanced Electrocardiography for Detection and Prediction of Hypertrophic Cardiomyopathy across Monogenic and Polygenic Susceptibility

The study evaluated the use of artificial intelligence in electrocardiogram analysis (AI-ECG) to detect hypertrophic cardiomyopathy (HCM) in 1,095 individuals with genetic variants increasing disease risk. The AI-ECG model achieved 91 percent accuracy in identifying individuals with HCM manifestations at first assessment and 92 percent in detecting manifest disease. Among individuals without disease manifestations at baseline, AI-ECG predicted HCM development during follow-up with a 1.55-fold higher risk per unit increase in score. In the UK Biobank population, individuals with both high AI-ECG score and high polygenic risk had 60-fold higher odds of HCM compared to 15-fold risk for AI-ECG alone. The results suggest that AI-ECG is a scalable tool for HCM detection and can guide surveillance in individuals with genetic disease risk.