Diagnostic performance of the ECG-SMART AI score for detecting angiographically significant coronary artery disease in patients undergoing elective coronary angiography
9 affiliations
- 1School of Nursing, Middle East University, Amman, Jordan
- 2School of Nursing, University of Jordan, Amman, Jordan
- 3Department of Cardiology, Jordan University Hospital, Amman, Jordan
- 4School of Nursing, University of Rochester, Rochester, New York, United States of America
- 5School of Engineering, Texas Tech University, Lubbock, Texas, United States of America
- 6Department of Electrical & Computer Engineering, University of Toronto, Ontario, Canada
- 7Division of Cardiology, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America
- 8INTELLICARDIO INC., Pittsburgh, Pennsylvania, United States of America
- 9Department of Emergency Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania, United States of America
Abstract
The 12-lead ECG is central to the evaluation of patients with suspected coronary artery disease (CAD) but has limited diagnostic accuracy for identifying significant obstructive disease prior to elective coronary angiography. We previously developed an artificial intelligence–enhanced ECG (AI-ECG) model for detecting acute coronary occlusion. Whether this model generalizes to identifying angiographically significant CAD in symptomatic patients undergoing elective evaluation remains unknown. In this prospective cohort study, symptomatic outpatients undergoing elective coronary angiography at a tertiary center received standard 12-lead ECGs before the procedure. ECGs were analyzed offline using a previously validated AI-ECG model originally developed for acute coronary occlusion. Patients were categorized as low or intermediate-to-high risk using predefined thresholds. The primary outcome was angiographically significant CAD (≥70% stenosis in a major epicardial vessel or ≥50% stenosis of the left main coronary artery). Among 363 patients (age 58.9 ± 11 years; 55.1% male), 36.1% had angiographically significant CAD. The AI-ECG classified 56% as low risk and 44% as intermediate-to-high risk. After adjustment for conventional clinical characteristics, intermediate-to-high risk remained independently associated with significant CAD (OR 3.12, 95% CI 1.88–5.20; p < 0.001). Model discrimination was good (AUROC 0.79, 95% CI 0.74–0.84), with 60% precision at 80% recall. In summary, the ECG-SMART AI score demonstrated good diagnostic performance for identifying angiographically significant CAD in symptomatic patients undergoing elective coronary angiography. These findings support the generalizability of a previously developed AI-ECG model beyond acute coronary occlusion and suggest that AI-enhanced ECG analysis may complement existing clinical assessment by providing additional noninvasive information to refine pretest probability during elective evaluation. Author summary: Coronary artery disease is one of the leading causes of illness and death worldwide. Identifying patients with important narrowing of the heart arteries can be challenging, particularly when symptoms are not acute. The electrocardiogram is inexpensive, widely available, and routinely used in cardiovascular care, but conventional interpretation has limited ability to detect underlying coronary artery narrowing in stable patients. We previously developed an artificial intelligence approach that analyzes information within a standard 12-lead electrocardiogram to identify patients at risk for an acutely blocked coronary artery. In this study, we asked whether the same approach could provide useful information in a very different setting: symptomatic patients undergoing planned coronary angiography rather than emergency evaluation for a heart attack. Among 363 patients, the artificial intelligence score successfully identified patients more likely to have significant coronary artery narrowing and remained informative after accounting for conventional cardiovascular risk factors. These findings suggest that artificial intelligence may uncover subtle electrical patterns associated with coronary disease that extend beyond acute heart attacks. Further studies are needed to determine whether this approach can help clinicians improve noninvasive evaluation and guide decisions about additional testing in broader patient populations.
Related
- Incremental diagnostic value of coronary CT angiography-derived pericoronary adipose tissue radiomics for identifying hemodynamically significant coronary stenosis
- Deep Learning Optimizes the Diagnostic Performance of Coronary Computed Tomography Angiography in Heavily Calcified Coronary Plaques: A Chinese Multicenter Study
- Diagnostic Performance of Triglyceride-Glucose Body Mass Index (TyG-BMI), Fibrosis-4 Index (FIB-4), and Their Combined Model for Clinically Significant Liver Fibrosis in Type 2 Diabetes Mellitus
- Microvascular imaging in diffuse thyroid disease: diagnostic performance of subtraction ultra-micro angiography
- Prognostic Value of Plaque Burden Beyond Stenosis Severity in Patients with Stable Chest Pain Undergoing Coronary CT Angiography
BibTeX
@article{hani2026diagnostic,
title = {{Diagnostic performance of the ECG-SMART AI score for detecting angiographically significant coronary artery disease in patients undergoing elective coronary angiography}},
author = {Dania A. Bani Hani and Jafar A. Alshraideh and Hamza Alduraidi and Akram Saleh and Nathan T. Riek and Karam Daoud and Rui Qi Ji and Samir Saba and Clifton W. Callaway and Salah Al-Zaiti},
journal = {PLOS Digital Health},
year = {2026},
month = oct,
doi = {10.1371/journal.pdig.0001701},
url = {https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001701},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Diagnostic performance of the ECG-SMART AI score for detecting angiographically significant coronary artery disease in patients undergoing elective coronary angiography
AU - Dania A. Bani Hani
AU - Jafar A. Alshraideh
AU - Hamza Alduraidi
AU - Akram Saleh
AU - Nathan T. Riek
AU - Karam Daoud
AU - Rui Qi Ji
AU - Samir Saba
AU - Clifton W. Callaway
AU - Salah Al-Zaiti
PY - 2026
DA - 2026/10/08
JO - PLOS Digital Health
DO - 10.1371/journal.pdig.0001701
UR - https://journals.plos.org/digitalhealth/article?id=10.1371/journal.pdig.0001701
AB - The 12-lead ECG is central to the evaluation of patients with suspected coronary artery disease (CAD) but has limited diagnostic accuracy for identifying significant obstructive disease prior to elective coronary angiography. We previously developed an artificial intelligence–enhanced ECG (AI-ECG) model for detecting acute coronary occlusion. Whether this model generalizes to identifying angiographically significant CAD in symptomatic patients undergoing elective evaluation remains unknown. In this prospective cohort study, symptomatic outpatients undergoing elective coronary angiography at a tertiary center received standard 12-lead ECGs before the procedure. ECGs were analyzed offline using a previously validated AI-ECG model originally developed for acute coronary occlusion. Patients were categorized as low or intermediate-to-high risk using predefined thresholds. The primary outcome was angiographically significant CAD (≥70% stenosis in a major epicardial vessel or ≥50% stenosis of the left main coronary artery). Among 363 patients (age 58.9 ± 11 years; 55.1% male), 36.1% had angiographically significant CAD. The AI-ECG classified 56% as low risk and 44% as intermediate-to-high risk. After adjustment for conventional clinical characteristics, intermediate-to-high risk remained independently associated with significant CAD (OR 3.12, 95% CI 1.88–5.20; p < 0.001). Model discrimination was good (AUROC 0.79, 95% CI 0.74–0.84), with 60% precision at 80% recall. In summary, the ECG-SMART AI score demonstrated good diagnostic performance for identifying angiographically significant CAD in symptomatic patients undergoing elective coronary angiography. These findings support the generalizability of a previously developed AI-ECG model beyond acute coronary occlusion and suggest that AI-enhanced ECG analysis may complement existing clinical assessment by providing additional noninvasive information to refine pretest probability during elective evaluation. Author summary: Coronary artery disease is one of the leading causes of illness and death worldwide. Identifying patients with important narrowing of the heart arteries can be challenging, particularly when symptoms are not acute. The electrocardiogram is inexpensive, widely available, and routinely used in cardiovascular care, but conventional interpretation has limited ability to detect underlying coronary artery narrowing in stable patients. We previously developed an artificial intelligence approach that analyzes information within a standard 12-lead electrocardiogram to identify patients at risk for an acutely blocked coronary artery. In this study, we asked whether the same approach could provide useful information in a very different setting: symptomatic patients undergoing planned coronary angiography rather than emergency evaluation for a heart attack. Among 363 patients, the artificial intelligence score successfully identified patients more likely to have significant coronary artery narrowing and remained informative after accounting for conventional cardiovascular risk factors. These findings suggest that artificial intelligence may uncover subtle electrical patterns associated with coronary disease that extend beyond acute heart attacks. Further studies are needed to determine whether this approach can help clinicians improve noninvasive evaluation and guide decisions about additional testing in broader patient populations.
SN - 2767-3170
ER -