Detection of Hypertrophic Cardiomyopathy Using a Convolutional Neural Network-Enabled Electrocardiogram

Area of Science:

• Cardiology
• Artificial Intelligence
• Medical Diagnostics

Background:

• Hypertrophic cardiomyopathy (HCM) is a significant, though uncommon, cause of sudden cardiac death.
• Early and accurate detection of HCM is crucial for patient management and risk stratification.

Purpose of the Study:

• To develop and validate an artificial intelligence (AI) approach for detecting hypertrophic cardiomyopathy (HCM) using 12-lead electrocardiography (ECG).

Main Methods:

• A convolutional neural network (CNN) was trained on ECG data from 2,448 HCM patients and 51,153 controls.
• The CNN model was validated and tested on separate datasets, evaluating its diagnostic performance using area under the curve (AUC), sensitivity, and specificity.

Main Results:

• The AI model achieved a high AUC of 0.96 in the testing dataset, with 87% sensitivity and 90% specificity.
• The model demonstrated strong performance across various subgroups, including patients with ECG-defined left ventricular hypertrophy and normal ECGs.
• Performance was particularly notable in younger patients (95% sensitivity, 92% specificity).

Conclusions:

• AI-driven ECG analysis can effectively detect HCM with high diagnostic performance.
• The developed AI model shows potential for HCM screening, particularly in younger populations, but requires further validation.