AI-ECG for One-Year Mortality Risk Prediction

An Artificial Intelligence-Based Electrocardiogram Analysis System for One-Year Mortality Risk Prediction

Cardiovascular disease is a leading cause of death globally. While electrocardiogram (ECG) is a standard and accessible tool for heart screening, traditional risk assessment models often rely heavily on blood tests, which might be missing in electronic health records. To address this, the "Chang Gung ECG Mortality Risk Prediction Software" was developed. This artificial intelligence (AI) device analyzes standard 10-second, 12-lead resting ECG signals to predict the probability of cardiac-related mortality within one year.This study is a multi-center, retrospective cohort study designed to validate the clinical performance of this AI software. Researchers will analyze historical ECG data from patients aged 20 and older with suspected heart diseases across three hospitals in Taiwan. The AI's risk predictions will be compared against the actual one-year mortality outcomes documented in the patients' medical records. The primary goal is to determine if the AI model can accurately and consistently stratify patients' risk of cardiac-related death (such as heart failure, arrhythmia, and myocardial infarction) with an Area Under the Curve (AUC) greater than 0.80. This software is intended to serve as a clinical decision-support tool for long-term risk stratification in non-acute clinical settings, ultimately assisting physicians in providing better patient management.

Study Overview

• Status: Completed
• Conditions: Electrocardiogram; Mortality Risk Prediction
• Intervention / Treatment: Device: Chang Gung ECG Mortality Risk Prediction Software (Model: CGMH-DRP-001)

Detailed Description

Cardiovascular disease (CVD) is a major global health burden. Current risk assessment models (e.g., Framingham, QRISK) rely heavily on blood biochemistry data, which limits their applicability when electronic health record (EHR) data is incomplete. The 12-lead resting electrocardiogram (ECG) is a rapid, non-invasive, and highly accessible screening tool. Recent advancements in artificial intelligence (AI), specifically deep learning networks (such as ResNet), have demonstrated superior automatic feature extraction capabilities from ECG signals for predicting CVD risks.

This national multi-center retrospective study aims to evaluate the efficacy of a standalone Medical Device Software (SaMD), the Chang Gung ECG Mortality Risk Prediction Software. The core algorithm utilizes a 1D-ResNet-18 convolutional neural network to analyze 10-second, 12-lead resting ECG digital signals (sampled at 500Hz with a 60Hz AC filter). The software outputs a one-year mortality risk probability related to cardiac conditions to assist physicians in non-acute clinical settings.

Study Methodology The study will retrospectively collect and de-identify electronic health records and ECG data (from August 2011 to September 2024) across three institutions in Taiwan: Tri-Service General Hospital, Kaohsiung Armed Forces General Hospital, and Taipei Municipal Wanfang Hospital. Only the first eligible ECG per patient is included to prevent intra-individual bias.

The AI model's predictions will be compared against the actual one-year mortality outcomes. To ensure interpretability, cardiologists with over 5 years of clinical experience will review high-risk predictions using Gradient-weighted Class Activation Mapping (Grad-CAM). A prediction is considered correct only if both the risk assessment and the Grad-CAM localization are clinically reasonable.

Statistical Analysis The study employs a one-tailed superiority design with a significance level of 0.05. The null hypothesis states that the Area Under the Receiver Operating Characteristic Curve (AUC) is ≤ 0.80, while the alternative hypothesis targets an AUC > 0.80. Subgroup analyses will be conducted based on age distributions (e.g., 20-40, 41-60, >60 years) and specific cardiac etiologies (e.g., arrhythmias, myocardial infarction, heart failure) using DeLong's test to evaluate the model's predictive performance across different demographic and clinical scenarios.

Data Privacy and Federated Learning All patient data will be strictly de-identified according to HIPAA guidelines and analyzed within closed, secure intra-hospital networks. If the initial validation fails to meet the expected performance (AUC > 0.80), a federated learning approach will be initiated using a horizontal architecture. A maximum of 10% of the dataset will be used for model fine-tuning, after which the updated model will be independently re-validated using untouched data to ensure robustness and prevent data contamination.

• Study Type: Observational
• Enrollment (Actual): 461982

Contacts and Locations

Study Locations

• Taiwan
  • Kaohsiung City, Taiwan, 807
    • Kaohsiung Armed Forces General Hospital
  • Taipei, Taiwan, 114
    • Tri-Service General Hospital
  • Taipei, Taiwan, 114
    • Taipei Municipal Wanfang Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Adult; Older Adult
• Accepts Healthy Volunteers: N/A

• Sampling Method: Non-Probability Sample

Study Population

The study population consists of adult patients (aged 20 years and older) with suspected heart diseases who underwent standard 12-lead resting ECG examinations. Data are retrospectively collected from three medical institutions in Taiwan (Tri-Service General Hospital, Kaohsiung Armed Forces General Hospital, and Taipei Municipal Wanfang Hospital) between August 2011 and September 2024. The population encompasses a diverse real-world demographic across different clinical settings, including outpatient clinics, inpatient wards, and emergency departments, with comprehensive documentation of clinical diagnoses and mortality outcomes.

Description

Inclusion Criteria:

• Adults aged 20 years and older.
• Patients who underwent a 12-lead resting electrocardiogram (ECG) examination.
• ECG data must strictly meet the software input specifications: 12-lead, 500Hz sampling rate, 60Hz AC filter, and 10-second duration in XML format.
• Only the first eligible 12-lead ECG record per patient is included to prevent intra-individual bias.

Exclusion Criteria:

• ECG records with missing leads.
• Cases with missing demographic information (e.g., age, gender, mortality status) or missing clinical diagnostic data.
• ECG records that do not meet the product input specifications (e.g., incorrect sampling rate, AC filter settings, duration, or non-XML format).
• Pregnant women, patients under twenty years old, and patients with implanted pacemakers.

Study Plan

How is the study designed?

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Measure Description	Time Frame
Area Under the Receiver Operating Characteristic Curve (AUC) for One-Year Cardiac-Related Mortality Risk	The primary outcome is to evaluate the predictive performance of the AI model for mortality specifically related to cardiac conditions (such as heart failure, arrhythmias, and myocardial infarction) within one year. The performance will be assessed using the Area Under the Curve (AUC) metric. The trial uses a superiority design with a predefined success threshold of AUC > 0.80. The AI predictions will be retrospectively compared against the actual clinical outcomes documented in the electronic health records and death registry.	Up to 1 year (365 days) from the index ECG examination.

Collaborators and Investigators

• Sponsor: National Defense Medical Center, Taiwan
• Collaborators: Taipei Medical University WanFang Hospital; Kaohsiung Armed Forces General Hospital
• Investigators: Principal Investigator: Chin Lin, PhD, National Defense Medical Center

Study record dates

Study Major Dates

• Study Start (Actual): April 1, 2025
• Primary Completion (Actual): July 21, 2025
• Study Completion (Actual): July 21, 2025

Study Registration Dates

• First Submitted: June 15, 2026
• First Submitted That Met QC Criteria: June 15, 2026
• First Posted (Actual): June 22, 2026

Study Record Updates

• Last Update Posted (Actual): June 22, 2026
• Last Update Submitted That Met QC Criteria: June 15, 2026
• Last Verified: June 1, 2026

More Information

Terms related to this study

• Other Study ID Numbers: A202503002

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: NO

Drug and device information, study documents

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