Extraction of Risk Markers from ECG in Patients with Hypertrophic Cardiomyopathy

Hypertrophic cardiomyopathy (HCM) represents one of the leading causes of sudden cardiac death (SCD), particularly in the young population, with a risk of approximately 1% per year. So far, no reliable electrocardiogram (ECG) biomarkers have been presented for risk assessment, but ECG in HCM patients are often abnormal due to structural and electrical abnormalities. This study aimed to extract morphological ECG biomarkers to differentiate HCM patients based on their arrhythmic risk levels (15 HCM patients with arrhythmic events vs. 40 HCM control). We extracted ECG features including width, amplitudes, slopes between fiducial points, Hermite transform coefficients, and variational mode decomposition features. Following feature selection using combined metrics, the study population was divided into two groups for each ECG biomarker, with the median value serving as the cutoff point to distinguish between the groups. QRS and T waverelated features effectively separated patients into high and low arrhythmic risk categories. Notably, univariate Cox regression analysis showed that patients having more local QRS optima or highest percentage of negative QRS present the highest risk (p< 0.01 and p< 0.05 respectively). In conclusion, we proposed automatic ECG extracted features that can be used to stratify the risk for arrhythmic events in HCM patients.Clinical Relevance - This study provides novel insights into ECG-based risk stratification for HCM patients, offering potential tools for early identification of individuals at higher risk of cardiac events.