AI-ECG score

1. AI-ECG score improves heart-failure risk discrimination in humans (0.775 (0.769-0.780) to 0.794 (0.788-0.800)).

   “Discrimination improved from 0.775 (0.769-0.780; baseline score only) to 0.794 (0.788-0.800; time-varying model).”

   Humanstime-varying model vs baseline score only

   Dynamic Artificial Intelligence-enhanced Electrocardiographic Trajectories Outperform Single-Time-Point Screening for New-onset Heart Failure

   AHA Scientific Sessions 2025 abstract·3 Nov 2025

1. AI-ECG score is associated with mortality in humans (HR 1.08 [95%CI: 1.01-1.15]).

   “AI-ECG score (HR 1.08 [95%CI: 1.01-1.15]) and DASSi (1.17 [1.10-1.25]), but not peak velocity (1.04 [0.98-1.10]) were associated with mortality”

   Humansmedian follow-up of 4.6 yearsn = 61,812

   Multimodal AI Signatures of Cardiac Remodeling Independently Predict Worse Aortic Stenosis Prognosis

   AHA Scientific Sessions 2025 abstract·3 Nov 2025

2. AI-ECG score raises the risk of mortality in humans (aHR 1.16 (95% CI 1.07–1.26, p<0.001)).

   “1.16 (1.07–1.26, p<0.001) in the UK biobank.”

   Humansadjusted for age, sex, and clinical comorbidities in UK biobankn = 43,347

   A multi-ethnic foundation model-based artificial intelligence electrocardiogram for detection and prognostication of elevated left ventricular filling pressure

   AHA Scientific Sessions 2025 abstract·3 Nov 2025

1. AI-ECG score predicts incident aortic valve replacement in humans (HR 2.56 [95%CI: 2.16-3.04] per 1 SD).

   “All 3 biomarkers independently predicted AS progression based on incident AVR (per 1 SD: 2.56 [95%CI: 2.16-3.04], 2.56 [2.03-3.25] and 1.61 [1.24-2.08], respectively)”

   Humansmedian follow-up of 4.6 yearsn = 61,812

   Multimodal AI Signatures of Cardiac Remodeling Independently Predict Worse Aortic Stenosis Prognosis

   AHA Scientific Sessions 2025 abstract·3 Nov 2025

1. AI-ECG score raises the risk of heart failure in humans (HR 1.51, 95% CI 1.49–1.52).

   “each 0.1-unit increase in the time-updated AI-ECG prediction was associated with a 51% higher risk of HF (HR 1.51, 95% CI 1.49–1.52), accounting for age and sex.”

   Humansadjusted for age and sex

   Dynamic Artificial Intelligence-enhanced Electrocardiographic Trajectories Outperform Single-Time-Point Screening for New-onset Heart Failure

   AHA Scientific Sessions 2025 abstract·3 Nov 2025

More on Heart failure

1. AI-ECG score predicts left ventricular diastolic dysfunction grade II/III in humans (AUROC 0.941 (95% CI 0.931–0.951)).

   “For detecting LVDD grade II/III, the model demonstrated an AUROCs of 0.901 (0.892–0.910) and 0.941 (0.931–0.951) in the internal and external test cohort, respectively.”

   Humansexternal test cohortn = 17,926

   A multi-ethnic foundation model-based artificial intelligence electrocardiogram for detection and prognostication of elevated left ventricular filling pressure

   AHA Scientific Sessions 2025 abstract·3 Nov 2025

More on Left ventricular diastolic dysfunction grade II/III

1. AI-ECG score predicts increased left ventricular filling pressure in humans (AUROC 0.868 (95% CI 0.859–0.877)).

   “The AI-ECG model detected increased LVFP with area under the receiver operating characteristic curves (AUROCs) of 0.868 (95% confidence interval, 0.859–0.877) and 0.850 (0.841–0.858) in the internal and external test cohort, respectively.”

   Humansinternal test cohortn = 9,278

   A multi-ethnic foundation model-based artificial intelligence electrocardiogram for detection and prognostication of elevated left ventricular filling pressure

   AHA Scientific Sessions 2025 abstract·3 Nov 2025