EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Artificial Intelligence-Enhanced Electrocardiography in the Cardiac Intensive Care Unit: Implications for Coronary Intervention Decision-Making

Jae Seok Bae, Jeong Yoon Jang, Yun-Ho Cho, Yujin Shin, Hyo Jin Lee, Jong-Hwa Ahn

Journal of Clinical Medicine · 11.11.2026 · 15 (22) · 8836

DOI: 10.3390/jcm15228836 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Clinical Medicine, online erschienen am 11.11.2026. Verfasst von Jae Seok Bae und 5 weiteren Autor:innen. Die Arbeit nennt 21 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Coronary intervention in the cardiac intensive care unit (CICU) demands rapid, accurate diagnosis and real-time hemodynamic monitoring to guide revascularization decisions. The 12-lead electrocardiogram (ECG) remains the most immediately available diagnostic tool in this setting; however, conventional interpretation is limited by inter-observer variability, ventilator-induced artifact, and inherent insensitivity for atypical coronary occlusion presentations. Artificial intelligence (AI)-powered ECG analysis has emerged as a promising adjunct to interventional cardiology practice, with the potential to support automated detection of occlusion myocardial infarction (OMI)—including cases missed by conventional STEMI criteria—while simultaneously providing multiparametric assessment of left ventricular function, hemodynamic status, and electrolyte derangements relevant to procedural safety. This narrative review synthesizes current evidence on the clinical utility of AI-ECG with direct relevance to coronary intervention decision-making, focusing on: (1) OMI and STEMI detection and the potential reduction in false catheterization laboratory activations; (2) pre- and peri-procedural risk stratification using AI-ECG-derived ventricular function and diastolic dysfunction parameters; (3) periprocedural arrhythmia monitoring; and (4) electrolyte derangement screening relevant to procedural safety. Composite illustrative scenarios based on the authors’ CICU experience describe how the ECG Buddy AI-ECG system (QCG platform, Korean Ministry of Food and Drug Safety-approved) may support the recognition of acute myocardial infarction (AMI) and the decision to proceed with percutaneous coronary intervention (PCI) in mechanically ventilated patients with septic shock—a scenario in which conventional diagnostic pathways are often confounded. These scenarios are presented for illustration only and do not constitute evidence of clinical benefit. Retrospective single-center studies in large unselected CICU populations (n > 9000, Mayo Clinic series) have shown that AI-ECG is independently associated with mortality beyond conventional echocardiographic parameters, supporting its potential role as a decision-support tool. Whether AI-ECG-assisted decision-making improves clinical outcomes in coronary intervention remains to be established in prospective studies.

Zitieren

APA

Bae, J. S., Jang, J. Y., Cho, Y., Shin, Y., Lee, H. J., & Ahn, J. (2026). Artificial Intelligence-Enhanced Electrocardiography in the Cardiac Intensive Care Unit: Implications for Coronary Intervention Decision-Making. Journal of Clinical Medicine, 15(22), 8836. https://doi.org/10.3390/jcm15228836

Vancouver

Bae JS, Jang JY, Cho Y, Shin Y, Lee HJ, Ahn J. Artificial Intelligence-Enhanced Electrocardiography in the Cardiac Intensive Care Unit: Implications for Coronary Intervention Decision-Making. Journal of Clinical Medicine. 2026;15(22):8836. doi:10.3390/jcm15228836

BibTeX
@article{Bae2026,
  author = {Bae, Jae Seok and Jang, Jeong Yoon and Cho, Yun-Ho and Shin, Yujin and Lee, Hyo Jin and Ahn, Jong-Hwa},
  title = {Artificial Intelligence-Enhanced Electrocardiography in the Cardiac Intensive Care Unit: Implications for Coronary Intervention Decision-Making},
  journal = {Journal of Clinical Medicine},
  year = {2026},
  volume = {15},
  number = {22},
  pages = {8836},
  doi = {10.3390/jcm15228836}
}
Kostenlos starten →