ETD

• Artificial Intelligence (AI)
• Cardiac magnetic resonance (CMR)
• DL
• ECG
• fibrosis
• Hypertrophic cardiomyopathy (HCM)

Hypertrophic cardiomyopathy (HCM) is one of the most prevalent inherited cardiac conditions, characterized by increased myocardial thickness and/or mass, and associated with a higher risk of mortality compared to the general population. In both suspected and confirmed cases of HCM, current European and American guidelines recommend the 12-lead electrocardiogram (ECG) as a first-line diagnostic tool and as part of longitudinal patient monitoring. ECGs in HCM patients often reveal features suggestive of left ventricular hypertrophy, including increased QRS voltage and repolarization abnormalities. Cardiac magnetic resonance (CMR) imaging also plays a central role in the diagnostic and prognostic workup of HCM, offering advanced tissue characterization and precise quantification of myocardial fibrosis through late gadolinium enhancement (LGE).
Over the past decade, artificial intelligence—particularly machine learning and deep learning techniques—has demonstrated remarkable versatility and clinical utility across various domains of cardiovascular medicine. Recent research has explored AI-enhanced ECG interpretation in diverse clinical scenarios, with a growing focus on cardiomyopathies. In the context of HCM, AI has shown potential not only in diagnosis and prognostication but also in predicting adverse events and therapeutic response using only standard 12-lead ECG data. The scope of our study was to investigate whether AI-enhanced ECG analysis could predict the presence of high risk imaging features in HCM, such as increased LV mass and fibrosis, thus improving patient selection and referral to higher level testing such as CMR.