Related Experiment Video
Updated: Jun 27, 2026
Testing Acetylcholine Followed by Adenosine for Invasive Diagnosis of Coronary Vasomotor Disorders
Published on: February 3, 2021
Development and Validation of Artificial Intelligence Prediction of Epicardial Coronary Artery Spasm in Patients
Ming-Jui Hung1, Ian Y Chen2, Yung-Neng Lin3
1Division of Cardiology, Department of Medicine, Chang Gung Memorial Hospital at Keelung, Chang Gung University College of Medicine, No. 222, Maijin Rd, Anle District, Keelung City 204201, Taiwan.
Insights
An artificial intelligence (AI) risk score accurately predicts coronary artery spasm (CAS) in patients without obstructive coronary artery disease (CAD). This AI tool aids in early identification and risk stratification for myocardial ischemia.
Area of Science:
- Cardiology
- Artificial Intelligence in Medicine
- Diagnostic Tools
Background:
- Epicardial coronary artery spasm (CAS) is a significant cause of myocardial ischemia.
- Non-obstructive coronary artery disease (CAD) can still present with CAS.
- Predicting CAS noninvasively is crucial for patient management.
Purpose of the Study:
- To develop and validate a noninvasive AI-driven risk score for predicting CAS.
- To assess the utility of machine learning models using clinical and electrocardiogram data for CAS prediction.
Main Methods:
- Retrospective analysis of 1050 patients (derivation) and 600 patients (validation) undergoing methylergonovine provocation testing.
- Development of a nine-variable Random Forest (RF) model using clinical data.
- Implementation of a CNN-LSTM deep learning model using raw electrocardiogram (ECG) data.
Main Results:
- The nine-variable RF model showed strong discriminatory power (AUC 85.8% derivation, 84.1% validation).
- A dose-response relationship was observed between risk factors and CAS prevalence.
- The ECG-based CNN-LSTM model demonstrated high sensitivity (91.4%) but low specificity (11.9%), serving as a proof of concept.
Conclusions:
- The nine-variable RF model is a practical and accurate tool for early CAS identification and risk stratification.
- The ECG deep learning model, while preliminary, has potential to complement the RF model for improved clinical decision-making.
Abstract:
Background: Epicardial coronary artery spasm (CAS) is a frequent and important cause of myocardial ischemia. We aimed to develop and validate a noninvasive, artificial intelligence (AI)-driven risk score using routine clinical data to predict CAS in patients without obstructive coronary artery disease (CAD). Methods: This retrospective study analyzed a derivation cohort of 1050 patients and an external validation cohort of 600 patients who underwent intracoronary methylergonovine provocation testing between September 2008 and March 2025. A random forest (RF) model was developed using 15 clinical variables and simplified to a nine-variable model. Additionally, a convolutional neural network-long short-term memory (CNN-LSTM) deep learning model was implemented to predict CAS from raw digital electrocardiogram data (2611 electrocardiogram records). Results: The final nine-variable RF model, including predictors such as diastolic/systolic blood pressure, age, BSA, hemoglobin, smoking, heart rate, sex, and estimated glomerular filtration rate, demonstrated strong discriminatory power. The area under the curve was 85.8% (95% confidence interval [CI]: 85.8-89.9%) in the derivation cohort and 84.1% in the validation cohort (95% CI: 80.6-87.7%). A dose-response relationship was confirmed, with CAS prevalence increasing from 42.1% (0-1 risk factors) to 82.4% (≥5 risk factors). The electrocardiogram-based CNN-LSTM deep learning model achieved high sensitivity (91.4%) but limited specificity (11.9%); therefore, it should be considered a proof of concept rather than a clinical screening tool until further refinement is achieved. Conclusions: The nine-variable RF model provides a practical and accurate tool for early identification and risk stratification of CAS. The electrocardiogram deep learning model complements the RF model to improve clinical decisions and resource allocation in diagnosing CAS.
