Artificial intelligence applied to electrocardiogram to rule out acute myocardial infarction: the ROMIAE multicentre study

Abstract

Background and aims: Emerging evidence supports artificial intelligence-enhanced electrocardiogram (AI-ECG) for detecting acute myocardial infarction (AMI), but real-world validation is needed. The aim of this study was to evaluate the performance of AI-ECG in detecting AMI in the emergency department (ED).

Methods: The Rule-Out acute Myocardial Infarction using Artificial intelligence Electrocardiogram analysis (ROMIAE) study is a prospective cohort study conducted in the Republic of Korea from March 2022 to October 2023, involving 18 university-level teaching hospitals. Adult patients presenting to the ED within 24 h of symptom onset concerning for AMI were assessed. Exposure included AI-ECG score, HEART score, GRACE 2.0 score, high-sensitivity troponin level, and Physician AMI score. The primary outcome was diagnosis of AMI during index admission, and the secondary outcome was 30 day major adverse cardiovascular event (MACE).

Results: The study population comprised 8493 adults, of whom 1586 (18.6%) were diagnosed with AMI. The area under the receiver operating characteristic curve for AI-ECG was 0.878 (95% CI, 0.868-0.888), comparable with the HEART score (0.877; 95% CI, 0.869-0.886) and superior to the GRACE 2.0 score, high-sensitivity troponin level, and Physician AMI score. For predicting 30 day MACE, AI-ECG (area under the receiver operating characteristic, 0.866; 95% CI, 0.856-0.877) performed comparably with the HEART score (0.858; 95% CI, 0.848-0.868). The integration of the AI-ECG improved risk stratification and AMI discrimination, with a net reclassification improvement of 19.6% (95% CI, 17.38-21.89) and a C-index of 0.926 (95% CI, 0.919-0.933), compared with the HEART score alone.

Conclusions: In this multicentre prospective study, the AI-ECG demonstrated diagnostic accuracy and predictive power for AMI and 30 day MACE, which was similar to or better than that of traditional risk stratification methods and ED physicians.

Keywords: AI/ML-enabled SaMD; Acute coronary syndrome; Acute myocardial infarction; Artificial intelligence; Electrocardiogram; Emergency department.

Structured Graphical Abstract

An overview of ROMIAE study including design and main findings. ROMIAE, Rule-Out acute Myocardial Infarction using Artificial intelligence Electrocardiogram analysis; AI-ECG, artificial intelligence–enhanced electrocardiogram; AMI, acute myocardial infarction; AUROC, area under the receiver operating characteristic curve; ED, emergency department; MACE, major adverse cardiovascular event; NPV, negative predictive value; PPV, positive predictive value.

Figure 1

Study flow. This illustrates the selection process for patient inclusion in this study, which was conducted in 18 emergency departments from March 2022 to October 2023

Figure 2

Primary and secondary outcome analyses using receiver operating characteristic curves. (A) Receiver operating characteristic curve for the primary outcome (diagnosis of acute myocardial infarction at the index visit). (B) Receiver operating characteristic curve for the secondary outcome (prediction of 30 day major adverse cardiovascular events). Each curve represents a different diagnostic tool or score: Artificial Intelligence–Enhanced Electrocardiogram score, HEART score, Physician Acute Myocardial Infarction score, initial high-sensitivity troponin level, and GRACE score. The area under the receiver operating characteristic curve for each diagnostic measure is displayed with the corresponding 95% confidence intervals, indicating the performance of each test in terms of its sensitivity (true positive rate) vs. 1−specificity (false positive rate). *P < .05

Figure 3

Risk stratification of acute myocardial infarction using Artificial Intelligence–Enhanced Electrocardiogram, hs-troponin, and HEART scores. (A) Classification of patients based on three acute myocardial infarction risk stratification tools: Artificial Intelligence–Enhanced Electrocardiogram, HEART score, and GRACE 2.0 score. Patients are categorized into low-, intermediate-, and high-risk groups for each tool. (B) Implementation scenario of Artificial Intelligence–Enhanced Electrocardiogram, evaluating patients based on initial Artificial Intelligence–Enhanced Electrocardiogram results, further stratified using high-sensitivity troponin levels and HEART scores. The HEART score is applied to patients initially classified as intermediate- or high-risk by Artificial Intelligence–Enhanced Electrocardiogram, further categorizing them into low-, intermediate-, high-, and very high-risk groups as depicted in the figure

Figure 3

Risk stratification of acute myocardial infarction using Artificial Intelligence–Enhanced Electrocardiogram, hs-troponin, and HEART scores. (A) Classification of patients based on three acute myocardial infarction risk stratification tools: Artificial Intelligence–Enhanced Electrocardiogram, HEART score, and GRACE 2.0 score. Patients are categorized into low-, intermediate-, and high-risk groups for each tool. (B) Implementation scenario of Artificial Intelligence–Enhanced Electrocardiogram, evaluating patients based on initial Artificial Intelligence–Enhanced Electrocardiogram results, further stratified using high-sensitivity troponin levels and HEART scores. The HEART score is applied to patients initially classified as intermediate- or high-risk by Artificial Intelligence–Enhanced Electrocardiogram, further categorizing them into low-, intermediate-, high-, and very high-risk groups as depicted in the figure