Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2025 Oct 13;14(20):7209.
doi: 10.3390/jcm14207209.

Deep Learning-Based Risk Assessment and Prediction of Cardiac Outcomes Using Single-Lead 24-Hour Holter-ECG in Patients with Heart Failure or Myocardial Infarction

Ju Youn Kim  1 , Kyung Geun Kim  2   3 , Sunghoon Joo  2 , Mineok Chang  2 , Juwon Kim  1 , Kyoung-Min Park  1 , Young Keun On  1 , June Soo Kim  1 , Young Soo Lee  4 , Seung-Jung Park  1 ; K-REDEFINE Investigators
Affiliations

Deep Learning-Based Risk Assessment and Prediction of Cardiac Outcomes Using Single-Lead 24-Hour Holter-ECG in Patients with Heart Failure or Myocardial Infarction

Ju Youn Kim et al. J Clin Med. .

Abstract

Background: Deep learning (DL) models using Holter-ECG may enhance risk stratification after heart failure (HF) or myocardial infarction (MI). Objective: To evaluate the prognostic performance of a Holter-based DL model for predicting major adverse cardiac events (MACE), compared with conventional noninvasive markers. Methods: In the K-REDEFINE study, 1108 patients with acute MI or HF underwent 24 h Holter monitoring. A DL model was trained using raw Holter-ECG data and tested for predicting a composite of cardiac death and ventricular arrhythmias. Its performance was compared with heart rate turbulence (HRT), T-wave alternans (TWA), and ejection fraction (EF). Results: During follow-up, 56 adjudicated cardiac deaths (1.18%/yr) and 21 ventricular arrhythmias (0.44%/yr) occurred. The DL model showed an area under the receiver operating characteristic curve (AUROC) of 0.74 (95% CI, 0.70-0.77) for the composite outcome, improving to 0.77 (0.74-0.81) when combined with EF. In comparison, HRT and TWA showed lower AUROCs of 0.62 and 0.55, respectively. For cardiac death alone, the AUROC reached 0.79, further improving to 0.82 with EF. Model-derived risk stratification revealed a seven-fold increase in cardiac death risk in the high-risk group compared to the low-risk group (HR 7.47, 95% CI 2.24-24.96, p < 0.001). This stratification remained particularly effective in patients with EF > 40%. Conclusions: A DL algorithm trained on single-lead Holter-ECG data effectively predicted cardiac death and ventricular arrhythmia. Its performance surpassed conventional markers and was further enhanced when integrated with EF, supporting its potential for noninvasive, scalable risk stratification.

Keywords: T-wave alternans; cardiac death; deep learning; ejection fraction; heart failure; heart rate turbulence; myocardial infarction; ventricular arrhythmia.

PubMed Disclaimer

Conflict of interest statement

Kyung Geun Kim and Mineok Chang were employees of VUNO Inc. during this work; Sunghoon Joo is an employee of VUNO Inc. All other authors declare no competing interests.

Figures

Figure 1
Figure 1
Receiver operating characteristic curves for the performance of the AI model and other variables for the composite of cardiac death and ventricular arrhythmia. (A) Total population (B) MI population (C) HF population.
Figure 2
Figure 2
Receiver operating characteristic curves for the performance of the AI model and other variables for cardiac death. (A) Total population (B) MI population (C) HF population.
Figure 3
Figure 3
Receiver operating characteristic curves for the performance of the AI model and other variables for composite of cardiac death and ventricular arrhythmia. (A) severely reduced left ventricular ejection fraction (B) preserved or mid-range reduced left ventricular ejection fraction.
Figure 4
Figure 4
Kaplan–Meier curve according to the risk group for cardiac death (A) Total population (B) preserved or mid-range reduced left ventricular ejection fraction (C) severely reduced left ventricular ejection fraction (Blue indicates low-risk, green indicates intermediate-risk, and yellow indicates high-risk group).
See this image and copyright information in PMC

References

    1. Pouleur A.C., Barkoudah E., Uno H., Skali H., Finn P.V., Zelenkofske S.L., Belenkov Y.N., Mareev V., Velazquez E.J., Rouleau J.L., et al. Pathogenesis of sudden unexpected death in a clinical trial of patients with myocardial infarction and left ventricular dysfunction, heart failure, or both. Circulation. 2010;122:597–602. doi: 10.1161/CIRCULATIONAHA.110.940619. - DOI - PubMed
    1. Zheng Z.J., Croft J.B., Giles W.H., Mensah G.A. Sudden cardiac death in the United States, 1989 to 1998. Circulation. 2001;104:2158–2163. doi: 10.1161/hc4301.098254. - DOI - PubMed
    1. Stecker E.C., Vickers C., Waltz J., Socoteanu C., John B.T., Mariani R., McAnulty J.H., Gunson K., Jui J., Chugh S.S. Population-Based Analysis of Sudden Cardiac Death with and Without Left Ventricular Systolic Dysfunction: Two-Year Findings from the Oregon Sudden Unexpected Death Study. J. Am. Coll. Cardiol. 2006;47:1161–1166. doi: 10.1016/j.jacc.2005.11.045. - DOI - PubMed
    1. Qu Z., Liu Q., Liu C. Classification of congestive heart failure with different New York Heart Association functional classes based on heart rate variability indices and machine learning. Expert Syst. 2019;36:e12396. doi: 10.1111/exsy.12396. - DOI
    1. Kwon J.-m., Kim K.-H., Jeon K.-H., Lee S.E., Lee H.-Y., Cho H.-J., Choi J.O., Jeon E.-S., Kim M.-S., Kim J.-J., et al. Artificial intelligence algorithm for predicting mortality of patients with acute heart failure. PLoS ONE. 2019;14:e0219302. doi: 10.1371/journal.pone.0219302. - DOI - PMC - PubMed

LinkOut - more resources