Synthetic Echo Motion May Aid ECG Risk Modeling

An ensemble electrocardiogram model classified derived diastolic dysfunction risk phenotypes and stratified heart failure–related death risk across external cohorts, according to findings presented at ASE 2026.

By
Kerri Miller
June 27, 2026
3 min

MDSpire News

Overview

A novel ensemble model integrating synthetic echocardiographic motion with ECG analysis effectively classified diastolic dysfunction risk phenotypes and stratified heart failure-related death risk. The model demonstrated strong predictive capabilities across multiple cohorts.

Background

Diastolic dysfunction is a critical aspect of heart failure, reflecting abnormalities in left heart physiology. Traditional echocardiography, while the gold standard for assessment, is not always feasible for widespread screening. This study explores an innovative approach to enhance risk stratification using electrocardiogram data.

Data Highlights

Cohort	AUC	4-Year Heart Failure-Related Death Rate
Development Cohort	0.86	-
External Test Cohort	0.85	-
EchoNext Cohort	0.74 to 0.83	-
CODE-15% Cohort	-	8.5% (high risk) vs 3.0% (low risk)

Key Findings

The ensemble model classified diastolic dysfunction risk phenotypes with AUCs of 0.86 and 0.85 in development and external cohorts, respectively.
Incremental predictive value was observed with a net reclassification improvement of 0.54 over the foundation ECG model.
High-risk ECG phenotypes correlated with structural remodeling indicators such as increased left ventricular mass index and left atrial volumes.
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 significantly higher 4-year heart failure-related death rate of 8.5% compared to 3.0% for low-risk patients.

Clinical Implications

The integration of synthetic echocardiographic motion into ECG analysis may enhance the detection of diastolic dysfunction and associated risks.

Conclusion

The findings indicate that combining echocardiographic risk states with synthetic cardiac motion in ECG models may improve the identification of cardiac abnormalities.