AI-Enhanced ECG flags aortic stenosis years before symptoms featuring AccurKardia
Matthew Segar, MD, of The Texas Heart Institute, discusses new research evaluating whether AI‑enabled ECG analysis can detect aortic stenosis years before traditional symptom‑driven diagnosis. Because echocardiography is effective but not scalable for population screening, his team applied the previously validated AccurKardia AK‑AVS model to community cohorts to determine whether subtle electrical remodeling—caused by chronic pressure overload—creates an ECG “fingerprint” detectable long before structural changes appear on echocardiography. Their findings show that elevated or rising AI ECG scores can precede TAVR by more than four years, accurately triage patients who need echocardiography, and identify “false positives” who actually carry substantially elevated long‑term risk for AS progression and heart failure. The work demonstrates how AI‑ECG trajectories reveal distinct biological pathways to severe AS and may guide earlier referral, surveillance frequency, and pre‑procedural optimization.
