Feasibility of Artificial Intelligence–Based Electrocardiography Analysis for the Prediction of Obstructive Coronary Artery Disease in Patients With Stable Angina: Validation Study

Abstract

Background Despite accumulating research on artificial intelligence–based electrocardiography (ECG) algorithms for predicting acute coronary syndrome (ACS), their application in stable angina is not well evaluated. Objective We evaluated the utility of an existing artificial intelligence–based quantitative electrocardiography (QCG) analyzer in stable angina and developed a new ECG biomarker more suitable for stable angina. Methods This single-center study comprised consecutive patients with stable angina. The independent and incremental value of QCG scores for coronary artery disease (CAD)–related conditions (ACS, myocardial injury, critical status, ST-elevation myocardial infarction, and left ventricular dysfunction) for predicting obstructive CAD confirmed by invasive angiography was examined. Additionally, ECG signals extracted by the QCG analyzer were used as input to develop a new QCG score. Results Among 723 patients with stable angina (median age 68 years; male: 470/723, 65%), 497 (69%) had obstructive CAD. QCG scores for ACS and myocardial injury were independently associated with obstructive CAD (odds ratio [OR] 1.09, 95% CI 1.03-1.17 and OR 1.08, 95% CI 1.02-1.16 per 10-point increase, respectively) but did not significantly improve prediction performance compared to clinical features. However, our new QCG score demonstrated better prediction performance for obstructive CAD (area under the receiver operating characteristic curve 0.802) than the original QCG scores, with incremental predictive value in combination with clinical features (area under the receiver operating characteristic curve 0.827 vs 0.730; P<.001). Conclusions QCG scores developed for acute conditions show limited performance in identifying obstructive CAD in stable angina. However, improvement in the QCG analyzer, through training on comprehensive ECG signals in patients with stable angina, is feasible.

Artificial intelligence–enhanced electrocardiography analysis as a promising tool for predicting obstructive coronary artery disease in patients with stable angina

“… 22 Given these advancements, we hypothesized that the AI-based ECG analysis could uncover the subtle or non-specific changes in stable angina, thereby improving obstructive CAD prediction. Our initial study on 723 stable angina patients introduced an AI model that utilizes deep learning-derived quantitative ECG features for obstructive CAD prediction, 11 showing superior performance to clinical risk factors. However, this initial study, similar to others, 23 , 24 was constrained by its small sample size and lack of external validation.…”

“…Electrocardiographies can be interpreted in various formats, ranging from photographs or printed outputs from conventional ECG recorders to digital ECG systems. 6 , 7 , 11 Its adaptability ensures that the QCG analyzer is suitable for a wide range of clinical environments, from advanced medical centers equipped with digital ECG systems to resource-limited settings that rely on paper-based ECGs.…”

“…The sigmoid outputs of the analyzer, namely, the QCG scores, were scaled between 0 and 100 for the presence of obstructive CAD (QCG ObstCAD ) and extensive CAD (QCG ExtCAD ). 6 , 7 , 11 Hyperparameter optimization was further introduced for the final model, including batch size and learning rate, using the data splitting technique within the training data set.…”

“… 6–10 However, our prior study deviated from this common focus by examining the potential of AI-based ECG analysis for predicting obstructive CAD in patients with stable angina. 11 The findings suggested that an AI-based ECG analysis can substantially improve the prediction of obstructive CAD, achieving superior accuracy over that with conventional clinical risk factors. Nonetheless, this initial algorithm was developed in a relatively small patient cohort and was not subjected to externally validation.…”

Artificial Intelligence–Based Electrocardiographic Biomarker for Outcome Prediction in Patients With Acute Heart Failure: Prospective Cohort Study

“…The tasks include the classification of 12 rhythms (with 35 subtypes) and production of 10 digital biomarkers correlated with the risk of (1) being critically ill (shock, respiratory failure, or cardiac arrest), (2) cardiac ischemia (acute coronary syndrome, ST-elevation myocardial infarction, or myocardial injury as defined by an elevated troponin level), (3) cardiac dysfunction (pulmonary edema, left and right heart dysfunction, pulmonary hypertension, and clinically significant pericardial effusion), and (4) hyperkalemia. Several validation studies of the system have been published previously [ 12 - 14 ]. The collection of these AI algorithms has been developed into a mobile app (ECG Buddy, ARPI), which has been approved by the Korean Ministry of Food and Drug Safety.…”