Extraction of Risk Markers from ECG in Patients with Hypertrophic Cardiomyopathy

Area of Science:

• Cardiology
• Biomedical Engineering
• Medical Diagnostics

Background:

• Hypertrophic cardiomyopathy (HCM) is a primary cause of sudden cardiac death (SCD) in young individuals, with an annual risk of ~1%.
• Electrocardiogram (ECG) abnormalities are common in HCM patients due to underlying structural and electrical changes, yet reliable ECG biomarkers for risk stratification are lacking.

Purpose of the Study:

• To extract novel morphological ECG biomarkers for differentiating HCM patients based on arrhythmic risk.
• To assess the efficacy of these biomarkers in stratifying HCM patients into high- and low-risk categories for arrhythmic events.

Main Methods:

• Extraction of various ECG features, including morphological characteristics (width, amplitudes, slopes) and advanced signal processing techniques (Hermite transform, variational mode decomposition).
• Feature selection using combined metrics, with median values as cutoffs to categorize patients.
• Univariate Cox regression analysis to identify significant risk predictors.

Main Results:

• QRS and T wave-related ECG features demonstrated effectiveness in separating HCM patients into high and low arrhythmic risk groups.
• Increased local QRS optima and a higher percentage of negative QRS were significantly associated with the highest risk of arrhythmic events (p<0.01 and p<0.05, respectively).

Conclusions:

• Automatic extraction of specific ECG features can effectively stratify hypertrophic cardiomyopathy patients according to their risk of arrhythmic events.
• These findings offer potential tools for early identification of high-risk individuals, aiding in the prevention of sudden cardiac death.