Objective:

To develop a model that integrates synthetic echocardiographic motion into ECG analysis for classifying diastolic dysfunction risk phenotypes and stratifying heart failure-related death risk.

Approach:

• Model Development: An ensemble model was created combining features from synthetic cardiac motion generated from 12-lead ECGs and embeddings from a foundation ECG model pretrained on over 10 million recordings.
• Training and Validation: The model was trained on a multicenter cohort of 1,012 patients and validated in an independent cohort of 956 patients, with additional validation in the EchoNext cohort of 100,000 patients and the CODE-15% cohort of 233,770 patients.

Key Findings:

• The ensemble model classified diastolic dysfunction risk phenotypes with an AUC of 0.86 in the development cohort and 0.85 in the external test cohort.
• High-risk ECG phenotypes were linked to structural remodeling and clinical risk factors such as age, hypertension, and chronic kidney disease.
• In the EchoNext cohort, the model identified structural heart diseases with AUCs ranging from 0.74 to 0.83.
• In the CODE-15% cohort, high-risk patients had a 4-year heart failure-related death rate of 8.5%, compared to 3.0% for low-risk patients.

Interpretation:

Combining echo-derived risk states with synthetic cardiac motion in ECG models may enhance detection of cardiac abnormalities and optimize echocardiography use in resource-limited settings.

Limitations:

• The abstract does not establish whether the model improves clinical outcomes.
• It does not assess the model's performance as a prospective screening tool.
• The potential to replace standard echocardiographic assessment is not evaluated.

Conclusion:

The integration of synthetic echo motion into ECG analysis may improve risk stratification for heart failure-related death.

Sources:

• Journal of the American Society of Echocardiography