Machine Learning-Based Prediction of Acute Mortality in Emergency Department Patients Using Twelve-Lead Electrocardiogram

Aug 4, 2026

Risk of acute mortality in emergency department (ED) patients remains high, and timely identification of high-risk individuals can improve care decisions. This study develops a machine-learning model that uses standard twelve-lead electrocardiograms (ECGs) to forecast acute mortality risk. A convolutional neural network (CNN) is trained with ECG tracings from 345,593 ED patients and evaluated across multiple mortality endpoints and causes of death. The model shows strong discrimination for 30-day mortality with an ROC area of 0.84 and supports reliable one-year risk prediction, serving as a potential screening complement to existing early warning scores.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Background

Methods

Data and model development

Results

Conclusions

The Texas Medical Center Library DigitalCommons@TMC | Faculty, Staff and Student Publications | McWilliams School of Biomedical Informatics | | --- | --- | | 1-1-2023
Machine Learning-Based Prediction of Acute Mortality in Emergency Department Patients Using Twelve-Lead Electrocardiogram
Po-Cheng Chang Zhi-Yong Liu Yu-Chang Huang
Yu-Chun Hsu
Jung-Sheng Chen
lelonweipsadfditdioitninl lohtr:s[https://digitalcommons.library.tmc.edu/uthshis_docs](https://digitalcommons.library.tmc.edu/uthshis_docs)
Part of the Bioinformatics Commons, Biomedical Informatics Commons, Cardiology Commons, Cardiovascular Diseases Commons, Data Science Commons, and the Emergency Medicine Commons | | Recommended Citation Chang, Po-Cheng; Liu, Zhi-Yong; Huang, Yu-Chang; Hsu, Yu-Chun; Chen, Jung-Sheng; Lin, Ching-Heng; Tsai, Richard; Chou, Chung-Chuan; Wen, Ming-Shien; Wo, Hung-Ta; Lee, Wen-Chen; Liu, Hao-Tien; Wang, ChunChieh; and Kuo, Chang-Fu, "Machine Learning-Based Prediction of Acute Mortality in Emergency Department Patients Using Twelve-Lead Electrocardiogram" (2023) . Faculty, Staff and Student Publications. 459. [https://digitalcommons.library.tmc.edu/uthshis_docs/459](https://digitalcommons.library.tmc.edu/uthshis_docs/459) This Article is brought to you for free and open access by the McWilliams School of Biomedical Informatics at DigitalCommons@TMC. It has been accepted for inclusion in Faculty, Staff and Student Publications by an authorized administrator of DigitalCommons@TMC. For more information, please contact [digcommons@library.tmc.edu](digcommons@library.tmc.edu). Authors Po-Cheng Chang, Zhi-Yong Liu, Yu-Chang Huang, Yu-Chun Hsu, Jung-Sheng Chen, Ching-Heng Lin, Richard Tsai, Chung-Chuan Chou, Ming-Shien Wen, Hung-Ta Wo, Wen-Chen Lee, Hao-Tien Liu, Chun-Chieh Wang, and Chang-Fu Kuo This article is available at DigitalCommons@TMC: [https://digitalcommons.library.tmc.edu/uthshis_docs/459](https://digitalcommons.library.tmc.edu/uthshis_docs/459) TYPE Original Research PUBLISHED 27 October 2023 DOI 10.3389/fcvm.2023.1245614 EDITED BY Bert Vandenberk, University Hospitals Leuven, Belgium REVIEWED BY Gau-Jun Tang, National Yang-Ming University, Taiwan Tsung-Chien Lu, National Taiwan University Hospital, Taiwan *CORRESPONDENCE Chun-Chieh Wang chcwang@ms17 . hinet. net Chang-Fu Kuo [zandis@gmail.com](zandis@gmail.com) RECEIVED 23 June 2023 ACCEPTED 13 October 2023 PUBLISHED 27 October 2023 CITATION Chang P-C, Liu Z-Y, Huang Y-C, Hsu Y-C, Chen J-S, Lin C-H, Tsai R, Chou C-C, Wen M-S, Wo H-T, Lee W-C, Liu H-T, Wang C-C and Kuo C-F (2023) Machine learning-based prediction of acute mortality in emergency department patients using twelve-lead electrocardiogram. Front. Cardiovasc. Med. 10:1245614 . doi: 10.3389/fcvm.2023.1245614 COPYRIGHT © 2023 Chang, Liu, Huang, Hsu, Chen, Lin, Tsai, Chou, Wen, Wo, Lee, Liu, Wang and Kuo. This isan open-access article distributed under the terms of the Creative Commons Attribution License (CC BY) . The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms. Machine learning-based prediction of acute mortality in emergency department patients using twelve-lead electrocardiogram Po-Cheng Chang1, Zhi-Yong Liu2, Yu-Chang Huang1, Yu-Chun Hsu2,3, Jung-Sheng Chen2, Ching-Heng Lin2, Richard Tsai2, Chung-Chuan Chou1, Ming-Shien Wen1, Hung-Ta Wo1, Wen-Chen Lee1, Hao-Tien Liu1, Chun-Chieh Wang1* and Chang-Fu Kuo2,4* 1Division of Cardiology, Department of Internal Medicine, Chang Gung Memorial Hospital, Linkou and Chang Gung University Medical School, Taoyuan, Taiwan, 2Center for Artiﬁcial Intelligence in Medicine, Chang Gung Memorial Hospital, Taoyuan, Taiwan, 3S

FAQs

What was the goal of the study?

To build a machine-learning model that uses the standard twelve-lead ECG to predict acute mortality risk in emergency department patients.

How was the machine-learning model developed and evaluated?

A convolutional neural network (CNN) ECG model was trained using ECG tracings from 345,593 ED patients, with patients split into training, validation, and testing datasets to assess mortality prediction performance.

What performance did the model achieve for 30-day mortality?

The model demonstrated high accuracy for acute mortality, with a hazard ratio of 8.50 and ROC curve areas of 0.84 for 30-day mortality risk prediction.

Aug 4, 2026

Upload by Theodore