What does the research say about ECG-predicted BNP?

ECG-predicted BNP

There are 2 papers on ECG-predicted BNP in aging research: none in people yet. The newest story, Nov 10, 2026: “AI predicts elevated heart-failure markers from heart recordings”.

Biomarker

2 papers study it3 findings

1 new paper in 30 daysSee all 2 papers on ECG-predicted BNP

“ECG-predicted BNP 100–399 was associated with a hazard ratio (HR) of 2.10 (95% CI: 1.74–2.53)”

Deep Learning-Based BNP Classification from Single-Lead ECGAHA Scientific Sessions 2025 abstract2025doi:10.1161/circ.152.suppl_3.4366733

In silico1

Associations

Heart failure

upin humans1HR 2.1

1 study

ECG-predicted BNP predicts heart failure in humans (prediction of incident HF in patients without preexisting HF) (c-index 0.74 [0.73-0.76] vs 0.70 [0.68-0.71], P<0.001).
“A model including age/sex/ECG-BNP (c-index 0.74 [0.73-0.76]) outperforms age/sex/BNP (c-index 0.70 [0.68-0.71]) for prediction of incident HF (P<0.001).”
Humansprediction of incident HF in patients without preexisting HFn = 6,278
Deep Learning-Based BNP Classification from Single-Lead ECG
AHA Scientific Sessions 2025 abstractNov 3, 2025
ECG-predicted BNP is linked with a higher risk of heart failure in humans (patients without prior HF) (HR 2.10 (95% CI: 1.74–2.53)).
“ECG-predicted BNP 100–399 was associated with a hazard ratio (HR) of 2.10 (95% CI: 1.74–2.53)”
Humanspatients without prior HFn = 6,278
Deep Learning-Based BNP Classification from Single-Lead ECG
AHA Scientific Sessions 2025 abstractNov 3, 2025

Latest

1 paper in 12 months

Journal of Cardiac FailureNov 10

AI predicts elevated heart-failure markers from heart recordings

Scores for distinguishing patients above and below heart-failure blood-marker thresholds ranged from 0.84 to 0.86 in a separate testing set.Scores for distinguishing patients above and below heart-failure blood-marker thresholds ranged from 0.84 to 0.86 in a separate testing set.

Computational studyHeart failure

Paper

AHA Scientific Sessions 2025 AbstractNov 3

Deep Learning-Based BNP Classification from Single-Lead ECG

Computational studyBiomarkers