AI-ECG for Echocardiography Triage in Structural Heart Disease: Evidence, Implementation, and Future Directions

Structural heart disease (SHD), including left ventricular systolic dysfunction, valvular heart disease, hypertrophic cardiomyopathy, cardiac amyloidosis, and pulmonary hypertension, remains underdiagnosed despite the increasing availability of disease-modifying therapies. Echocardiography is the principal confirmatory test, but its broad use as a screening tool is constrained by imaging capacity, cost, and referral efficiency. This review evaluates artificial intelligence-enabled 12-lead electrocardiography (AI-ECG) as a pre-echocardiographic triage tool for SHD. We synthesize evidence across reduced left ventricular ejection fraction, valvular disease, hypertrophic cardiomyopathy, cardiac amyloidosis, pulmonary hypertension, and composite SHD models, and distinguish two intended-use orientations: safety-net screening, in which a positive AI-ECG result serves as an additive trigger for confirmatory evaluation, and gatekeeper triage, in which a negative or low-risk AI-ECG result may support deferring or de-prioritizing echocardiography in selected low-risk settings. Current evidence most strongly supports low-LVEF detection, where pragmatic randomized implementation and early economic data are available. Valvular and composite SHD models are promising for referral enrichment, whereas hypertrophic cardiomyopathy, cardiac amyloidosis, and pulmonary hypertension remain earlier or pathway-incomplete applications. We also review false-positive interpretation, stepwise confirmation with point-of-care ultrasound, threshold selection, workflow integration, equity, regulation, and health economics. Overall, AI-ECG is currently best positioned as an additive safety-net tool to improve case finding upstream of echocardiography. Gatekeeper use remains investigational and requires prospective pathway-level validation, calibration, and operational safeguards before routine imaging deferral can be justified.