Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
February 8, 2026European Heart Journal

Detection of hospital-admitted heart failure regardless of ejection fraction using artificial intelligence-enhanced electrocardiograms

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Heart failure diagnosis is challenging, and previous AI-ECG models relied on echocardiographic labelling and often focused on specific ejection fraction phenotypes rather than identifying heart failure across the ejection fraction spectrum.

Does an AI-enhanced ECG model trained on ICD-10 codes and NT-proBNP levels accurately detect hospital-admitted heart failure regardless of ejection fraction?

Population

83,000 patients from a network of four tertiary hospitals

Comparison

AI-ECG model trained with ICD-10 and NT-proBNP labelling vs NT-proBNP levels and H2FPEF score

Design

Multicenter AI model development and prospective testing study

Key result

AI-ECG detected hospital-admitted heart failure with AUC 0.84 regardless of ejection fraction, outperforming NT-proBNP and identifying HFpEF even with low NT-proBNP.

Authors

ESElias StenhedeJRJesper RavnHSHenrik Schirmer

Discussion

Loading...

Member takes

Overview

May support ECG-based HF screening across EF; hypothesis-generating pending prospective validation.

Key Points

  • The research aims to create an AI model for detecting heart failure (HF) regardless of ejection fraction (EF) using electrocardiograms (ECGs).
  • Developed AI models trained on ECG data from 40,000 patients across four hospitals.
  • Used ICD-10 codes and NT-proBNP levels for labeling ECGs as HF or non-HF.
  • Evaluated model performance on a prospective test set of 43,000 patients with diagnosed HF.
  • Achieved an AUC of 0.84 for HF detection, outperforming NT-proBNP levels in diagnostic accuracy.
  • High diagnostic accuracy for HF with reduced EF (AUC of 0.91) and preserved EF (AUC ranging from 0.68 to 0.89).
  • Identified 24 out of 30 highest-risk patients meeting HFpEF criteria in a retrospective evaluation.

Structured PICO

Does an AI-enhanced ECG model trained on ICD-10 codes and NT-proBNP levels accurately detect hospital-admitted heart failure regardless of ejection fraction?

P
Population
83,000 patients from a network of four tertiary hospitals (40,000 patients in the development set from 2016-2022; 43,000 new patients in the prospective testing set from 2023-2024)
I
Intervention
Artificial intelligence-enhanced electrocardiogram (AI-ECG) model trained using ICD-10 codes and NT-proBNP levels without explicit echocardiographic labeling
C
Comparator
NT-proBNP levels and standard clinical diagnosis
O
Outcome
Diagnostic accuracy (AUC) for hospital-diagnosed heart failure regardless of ejection fraction and NT-proBNP levelssurrogate

An AI-enhanced ECG model trained without explicit echocardiographic labeling can accurately detect heart failure across the ejection fraction spectrum, potentially serving as an accessible screening tool in primary care.

Cite This Study

Stenhede et al. (2025) studied this question. AI-ECG detected hospital-admitted heart failure with AUC 0.84 regardless of ejection fraction, outperforming NT-proBNP and identifying HFpEF even with low NT-proBNP.

synapsesocial.com/papers/698828990fc35cd7a884835dhttps://doi.org/10.1093/eurheartj/ehaf784.1167
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Heart failure detection in electrocardiograms using Artificial Intelligence and pragmatic labelling2026
  2. 2Artificial Intelligence-Enabled Electrocardiogram Model for Predicting Heart Failure with Preserved Ejection Fraction– A Single-Center Study2025 · 10 citations
  3. 3An Artificial Intelligence Model for ECG-Based Prediction of Heart Failure with Preserved Ejection Fraction Diagnosis2026
  4. 4Electrocardiogram-based artificial intelligence to predict incident heart failure risk2026
  5. 5Artificial Intelligence-Enabled ECG Analysis to Predict Incident Heart Failure2026 · 2 citations