Artificial Intelligence Enhanced
Electrocardiography to Detect Occlusion Myocardial Infarction
in Acute Coronary Syndrome
Johnson Francis
Editor-in-Chief, BMH Medical Journal
Address for Correspondence: Johnson Francis, MBBS, MD,
DM, Editor-in-Chief, BMH Medical Journal, Baby Memorial
Hospital, Kozhikode, Kerala, India. E- mail: bmhmj0@gmail.com
Abstract
Coronary angiogram and subsequent percutaneous coronary
intervention are critical in the management of acute coronary
occlusions. The emerging paradigm distinguishing occlusion
myocardial infarction from non-occlusion myocardial infarction
will enhance diagnostic accuracy and prognostic effect in the
management of acute coronary syndrome. Artificial intelligence
enhanced electrocardiography can recognize subtle, acute
coronary occlusions which are often missed by the traditional
criteria. This can lead to timely cath lab activation,
enhanced myocardial salvage and better prognosis for patients
with myocardial infarction.
Acute coronary syndrome (ACS) includes ST elevation myocardial
infarction (STEMI), non-ST elevation myocardial infarction
(NSTEMI) and unstable angina. Coronary angiogram and
subsequent percutaneous coronary intervention are very
important in the management of acute coronary occlusions. The
STEMI criteria are electrocardiographic features which are
used to triage patients to catheterization laboratories.
Patients who do not fulfill STEMI criteria are triaged as
NSTEMIs. The NSTEMIs are managed with delayed coronary
reperfusion strategies. But a significant proportion of
patients with NSTEMI ACS are found to have acute occlusion of
coronary vessels and they need urgent reperfusion [1].
Adherence to STEMI/NSTEMI-based management algorithms will
delay reperfusion in these individuals, leading to increased
morbidity and mortality.
The limitations in the STEMI/NSTEMI framework underscore the
necessity for a revised approach to ACS management. The
emerging paradigm distinguishing occlusion myocardial
infarction (OMI) from non-occlusion myocardial infarction
(NOMI) will enhance diagnostic accuracy and prognostic effect
in the management of ACS. This can lead to earlier detection
of acute coronary occlusion, timely cath lab activation and
enhanced myocardial salvage [2].
The OMI paradigm shift uses clinical features,
electrocardiogram (ECG), Point-of-Care Ultrasound findings and
judicious use of troponin to find out patients in need of
emergent reperfusion. OMI ECG signs have high specificity and
can be learned by artificial intelligence (AI) [3]. The
OMI-NOMI paradigm offers greater prognostic precision for
patients with chest pain who require emergent angiography,
compared to the traditional STEMI-NSTEMI framework [4].
Artificial Intelligence Enhanced Electrocardiography in OMI
The 12-lead electrocardiogram is the common diagnostic test
for ischemic heart disease. The interpretation of ECG is
limited by human variability. Conventional ECG interpretation
often fails to detect subtle ischemic changes. This highlights
the necessity for artificial intelligence-driven approaches.
Through machine learning and artificial intelligence
algorithms, ECG analysis can be made more efficient. AI
enhanced ECG algorithms will detect myocardial infarction with
high accuracy and door-to-balloon coronary intervention time
can be reduced in ischemic heart disease [5].
Integrating AI in prehospital ECG interpretation improves the
identification of patients at low risk for having NSTEMI [6].
Artificial intelligence has transformed the management of ACS
by shifting the diagnostic focus to the OMI versus NOMI
paradigm. AI enhanced electrocardiography can recognize
subtle, acute coronary occlusions which are often missed by
the STEMI versus NSTEMI traditional criteria. AI enhanced ECG
enable earlier detection and precision-guided interventions in
occlusion myocardial infarction. Convolutional neural network
(CNN) and integrated neural network (INN)-enhanced ECGs
demonstrated superior performance in detecting acute coronary
syndrome with subtle ischemic ECG changes [7]. While
evaluating the performance of AI device in assessing
angiographic occlusion myocardial infarction and predicting
clinical outcomes, it was found that the AI model demonstrated
higher sensitivity compared with traditional STEMI criteria,
improving OMI detection [8].
AI models learn from large datasets and can identify complex
relationships that may not be obvious to human interpreters.
Deep learning techniques have improved the OMI diagnosis by
identifying subtle ECG patterns invisible to the human eye.
AI-enhanced ECG technology holds high potential for reshaping
cardiovascular diagnostics [9]. In high-risk NSTEMI, AI-ECG
model help to accurately diagnose OMI, enhancing early
identification of patients who require urgent invasive
coronary angiography [10]. AI tools which identify OMI earlier
than standard ECG interpretation can expedite intervention and
shorten reperfusion times.
DIFOCCULT-3 Trial (Time for a DIagnostic paradigm shift from
ST-elevation/non-ST-elevation to OCClUsion/non-occLusion
myocardial infarction) [11] design published in JACC: Advances
in October 2025 is evaluating this aspect. DIFOCCULT-3 is a
multicenter, modified cluster-randomized, open-label study
designed to compare the OMI/NOMI paradigm with the
STEMI/NSTEMI approach in 6,000 participants presenting with
suspected myocardial infarction. Study enrollment and
follow-up are ongoing. Acute outcomes will be reported after
enrolment and late outcomes including mortality and
rehospitalization will be published after completion of one
year follow-up.
Conclusion
AI-enhanced electrocardiography has enormous potential to
improve the management of cardiovascular disorders.
Integrating Artificial Intelligence with electrocardiograms
represents a transformative shift in in the management of ACS.
The occlusion myocardial infarction paradigm is based on the
presence or absence of acute coronary occlusion in the patient
rather than ST elevation on ECG. The OMI paradigm shift
harnesses advanced ECG interpretation augmented by artificial
intelligence. Artificial intelligence augmentation of ECG
assessment has significant role in improving the patient
outcomes in acute coronary syndrome. AI has shown promise in
detecting subtle ECG changes indicative of OMI, improving
diagnostic accuracy. Transitioning to the OMI paradigm and
emphasizing the identification of acute coronary occlusion
through AI-enhanced electrocardiography will significantly
improve patient outcomes by reducing delays in reperfusion
therapy.
References
1. Tudino R, Richardson E, Gonzalez JM, Barghout M, Ho T,
Abbott JD, Saad M. Acute coronary-occlusion myocardial
infarctions: performance of the STEMI criteria and evolving
alternative approaches. Trends Cardiovasc Med. 2026
Apr;36(3):180-187. doi: 10.1016/j.tcm.2025.10.006. Epub 2025
Oct 19. PMID: 41120030.
2. Ricci F, Martini C, Scordo DM, Rossi D, Gallina S,
Fedorowski A, Sciarra L, Chahal CAA, Meyers HP, Herman R,
Smith SW. ECG Patterns of Occlusion Myocardial Infarction: A
Narrative Review. Ann Emerg Med. 2025 Apr;85(4):330-340. doi:
10.1016/j.annemergmed.2024.11.019. Epub 2025 Jan 17. PMID:
39818676.
3. McLaren JTT, Nunes de Alencar J, Smith SW. Diagnosis of
occlusion myocardial infarction. Curr Opin Crit Care. 2026 Jun
1;32(3):230-238. doi: 10.1097/MCC.0000000000001359. Epub 2026
Jan 19. PMID: 41581009.
4. Umit TB, Sogut O, Arslan M, Az A, Dogan Y, Polat SN, Belen
E. Diagnostic challenges in acute coronary syndrome:
reconciling the STEMI-NSTEMI and OMI-NOMI paradigms. BMC
Cardiovasc Disord. 2026 Feb 10;26(1):228. doi:
10.1186/s12872-026-05612-3. PMID: 41667977; PMCID:
PMC12990446.
5. Kim EJ, Gala D, Ayyad M, Pramanik M, Makaryus AN. AI
Applications in Electrocardiography for Ischemic and
Structural Heart Disease: A Review of the Current State. J
Clin Med. 2026 Jan 1;15(1):316. doi: 10.3390/jcm15010316.
PMID: 41517565; PMCID: PMC12786525.
6. Demandt JPA, Mast TP, van Beek KAJ, Koks A, Bastiaansen
MCV, Tonino PAL, van 't Veer M, Zimmermann FM, Vlaar PJ.
Towards prehospital risk stratification using deep learning
for ECG interpretation in suspected acute coronary syndrome.
BMJ Health Care Inform. 2025 Jun Abstract6;32(1):e101292. doi:
10.1136/bmjhci-2024-101292. PMID: 40480678; PMCID:
PMC12161418.
7. Hori K, Suzuki S, Hirota N, Motogi J, Umemoto T, Nakai H,
Matsuzawa W, Takayanagi T, Hyodo A, Satoh K, Arita T, Yagi N,
Kishi M, Kano H, Matsuno S, Kato Y, Otsuka T, Uejima T, Yajima
J, Okumura Y, Oikawa Y, Yamashita T. Performance of
convolutional neural network-enhanced electrocardiography in
detecting acute coronary syndrome: focusing on subtypes and
reduced leads. J Cardiol. 2025 Sep;86(3):301-311. doi:
10.1016/j.jjcc.2025.05.014. Epub 2025 May 29. PMID: 40447041.
8. Choi JWH, Torelli V, Silverman A, Diaz SS, Kong D, Vaish E,
Katic L, Nagourney A, Khan Z, Robbins L, Pinney S, Barman N,
Farhan S. AI-enhanced recognition of occlusions in acute
coronary syndrome (AERO-ACS): a retrospective study. Coron
Artery Dis. 2026 Jan 1;37(1):39-45. doi:
10.1097/MCA.0000000000001555. Epub 2025 Aug 7. PMID: 40705383.
9. Palermi S, Vecchiato M, Ng FS, Attia Z, Cho Y, Anselmino M,
De Ferrari GM, Saglietto A; International AI-ECG Working
Group. Artificial intelligence and the electrocardiogram: A
modern renaissance. Eur J Intern Med. 2025 Oct;140:106329.
doi: 10.1016/j.ejim.2025.04.036. Epub 2025 May 20. PMID:
40413058.
10. Carvalho PEP, Belzer W, Pollmann DL, Helseth HC, Traverse
JH, Herman R, Mutlu D, Poulose A, Mooney M, Burke MN, Voudris
K, Goessl M, Chavez I, Wang YL, Strepkos D, Alexandrou M,
Cravero E, Rangan BV, Jaffe AS, Brilakis ES, Sharkey SW,
Sandoval Y. AI-Enhanced Electrocardiogram for Detection of
Occlusive Myocardial Infarction in High-Risk Non-ST-Segment
Elevation Acute Coronary Syndrome. JACC Adv. 2026
Apr;5(4):102663. doi: 10.1016/j.jacadv.2026.102663. Epub 2026
Mar 25. PMID: 41887018; PMCID: PMC13053
11. Aslanger EK, Aggul B, Yıldırımturk O, Karabay CY, Meyers
HP, Smith SW, Değertekin M; DIFOCCULT-3 Study Investigators. A
Diagnostic Paradigm Shift in Acute Myocardial Infarction:
Rationale and Design of the DIFOCCULT-3 Trial. JACC Adv. 2025
Nov;4(11 Pt 2):102227. doi: 10.1016/j.jacadv.2025.102227. Epub
2025 Oct 22. PMID: 41128712; PMCID: PMC12717606.