AI-ECG for Time-Resolved Prediction of HFrEF

Electrocardiogram-Based Deep Learning for Time-Resolved Prediction of Heart Failure With Reduced Ejection Fraction: A Multinational Study

This study aims to develop and validate a deep learning-based electrocardiogram (ECG) model for predicting the future risk of heart failure with reduced ejection fraction (HFrEF). The model is trained using raw 12-lead ECG data and generates individualized, time-resolved risk estimates over a 5-year period.

Data are obtained from multiple cohorts, including Zhongshan Hospital, Shanghai Tenth People's Hospital, and Beth Israel Deaconess Medical Center, representing diverse populations across China and the United States. The model is designed to identify individuals at elevated risk of developing HFrEF before the onset of overt clinical disease.

The performance of the model is evaluated using multiple complementary metrics, including discrimination, calibration, and clinical utility. In addition, interpretability analyses are conducted to explore the physiological relevance of ECG features associated with predicted risk.

This study seeks to provide an accessible and scalable tool for early risk stratification of heart failure, with the potential to support timely clinical decision-making and improve patient outcomes.

Study Overview

• Status: Completed
• Conditions: Heart Failure; Heart Failure With Reduced Ejection Fraction; Left Ventricular Dysfunction; Cardiac Remodeling

Detailed Description

Heart failure with reduced ejection fraction (HFrEF) is associated with substantial morbidity and mortality worldwide, and early identification of individuals at risk remains a major clinical challenge. Although existing risk models and biomarkers can provide prognostic information, their application is often limited by the need for laboratory testing or imaging, as well as variability in performance across populations.

In this study, we develop a deep learning-based survival model using raw 12-lead electrocardiogram (ECG) data to predict the future onset of HFrEF. The model is designed to generate individualized, time-to-event risk estimates over a 5-year follow-up period, allowing for dynamic assessment of risk trajectories rather than static classification.

The model is trained on data from Zhongshan Hospital and externally validated in independent cohorts from Shanghai Tenth People's Hospital and Beth Israel Deaconess Medical Center. These cohorts include a broad spectrum of patients, ranging from individuals without known cardiovascular disease to those with diverse clinical conditions, thereby enabling evaluation of model generalizability across different healthcare systems and demographic subgroups.

Model performance is comprehensively assessed using multiple metrics, including the concordance index, time-dependent area under the receiver operating characteristic curve, area under the precision-recall curve, Brier score, calibration analysis, and decision curve analysis. Risk stratification capability is evaluated using Kaplan-Meier survival analysis.

To enhance interpretability, complementary representation-based and attention-based methods are applied. These include variational autoencoder-derived latent feature analysis, correlation with conventional ECG parameters, and gradient-based visualization techniques to identify waveform regions contributing to model predictions. These approaches aim to ensure that the model captures physiologically meaningful signals associated with myocardial remodeling and cardiac dysfunction.

This study is observational and retrospective in nature and does not involve any intervention. The findings aim to support the development of a non-invasive, cost-effective, and widely accessible tool for early detection of individuals at risk of HFrEF, with potential implications for preventive strategies and personalized clinical management.

• Study Type: Observational
• Enrollment (Actual): 286709

Contacts and Locations

Study Locations

• China
  • Shanghai Municipality
    • Shanghai, Shanghai Municipality, China, 200436

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: Yes

• Sampling Method: Non-Probability Sample

Study Population

Participants were derived from three independent cohorts, including Zhongshan Hospital, Shanghai Tenth People's Hospital, and Beth Israel Deaconess Medical Center. The study population included adult patients who underwent routine ECG and echocardiographic evaluation in real-world clinical settings.

Description

Inclusion Criteria:

Adults aged ≥18 years Underwent standard 12-lead electrocardiography (ECG) Underwent transthoracic echocardiography with available LVEF measurement Availability of paired ECG-echocardiography data Data available for follow-up assessment

Exclusion Criteria:

Missing or incomplete ECG or echocardiography data Poor-quality ECG recordings unsuitable for analysis Missing key clinical variables required for model development

Study Plan

How is the study designed?

Number of groups / cohorts

1

Cohorts and Interventions

Group / Cohort

Overall Study Population; Participants from three independent cohorts (Zhongshan Hospital, Shanghai Tenth People's Hospital, and Beth Israel Deaconess Medical Center) who underwent standard 12-lead electrocardiography and echocardiographic evaluation. These data were used to develop and externally validate a deep learning model for time-to-event prediction of incident heart failure with reduced ejection fraction (HFrEF). No interventions were assigned, as this was an observational study based on routinely collected clinical data.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Incident Heart Failure With Reduced Ejection Fraction (HFrEF)	Occurrence of heart failure with reduced ejection fraction (HFrEF), defined as a left ventricular ejection fraction (LVEF) ≤40% during follow-up, as determined by transthoracic echocardiography. Both prevalent and incident cases identified from ECG-echocardiography data are included.	Up to 5 years

Collaborators and Investigators

• Sponsor: Shanghai Zhongshan Hospital
• Collaborators: Beth Israel Deaconess Medical Center; Shanghai 10th People's Hospital

Study record dates

Study Major Dates

• Study Start (Actual): February 1, 2014
• Primary Completion (Actual): December 1, 2023
• Study Completion (Actual): December 1, 2023

Study Registration Dates

• First Submitted: April 2, 2026
• First Submitted That Met QC Criteria: April 2, 2026
• First Posted (Actual): April 9, 2026

Study Record Updates

• Last Update Posted (Actual): April 9, 2026
• Last Update Submitted That Met QC Criteria: April 2, 2026
• Last Verified: April 1, 2026

More Information

Terms related to this study

• Keywords: Artificial Intelligence; Risk Stratification; Electrocardiogram; Deep Learning; Survival Model; Time-to-Event Prediction; Heart Failure Prediction; ECG-based Prediction
• Additional Relevant MeSH Terms: Cardiovascular Diseases; Heart Diseases; Ventricular Dysfunction; Heart Failure; Ventricular Dysfunction, Left
• Other Study ID Numbers: ZS-HF-AIECG-2026

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO
• IPD Plan Description: This study uses retrospective clinical data that are not publicly shareable.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No