Key result
ECG-based ML model predicting myocardial flow reserve <2 linked to ~285% higher major cardiovascular events risk.
Why the study?
Cardiac PET-derived myocardial flow reserve provides prognostic information but is expensive and selectively available, whereas ECG is inexpensive and ubiquitous.
Does a machine learning model using 12-lead ECG to predict myocardial blood flow reserve improve the prediction of major adverse cardiovascular events compared to baseline clinical models?
Observational (n=21,288)
Does a machine learning model using 12-lead ECG to predict myocardial blood flow reserve improve the prediction of major adverse cardiovascular events compared to baseline clinical models?
An AI model using standard 12-lead ECGs can accurately predict myocardial flow reserve and strongly stratify the risk of major adverse cardiovascular events, offering an accessible alternative to costly PET scans.
Answers are grounded in this paper and cite their sources.
May enhance noninvasive MACE stratification via ECG; hypothesis-generating and requires prospective validation before practice change.
We develop a machine learning (ML) model using electrocardiography (ECG) to predict myocardial blood flow reserve (MFR) and assess its prognostic value for major adverse cardiovascular events (MACEs). Using 3,639 ECG-positron emission tomography (PET) and 17,649 ECG-single-photon emission computed tomography (SPECT) data pairs, the ML model is trained with a swarm intelligence approach and support vector regression (SVR). The model achieves a receiver-operator curve (ROC) area under the curve (AUC) of 0.83, with a sensitivity and specificity of 0.75. An ECG-MFR value below 2 is significantly associated with MACE, with hazard ratios (HRs) of 3.85 and 3.70 in the discovery and validation phases, respectively. The model’s C-statistic is 0.76, with a net reclassification improvement (NRI) of 0.35. Validated in an independent cohort, the ML model using ECG data offers superior MACE prediction compared to baseline clinical models, highlighting its potential for risk stratification in patients with coronary artery disease (CAD) using the accessible 12-lead ECG. • Cardiac PET-derived myocardial flow reserve conveys highly prognostic information • Cardiac PET is expensive and selectively available; ECG is inexpensive and ubiquitous • Developed an ML model to predict MFR from ECG, validated on a large separate dataset • ECG-derived myocardial flow reserve is highly prognostic Alahdab et al. built an AI model using ECGs to predict myocardial flow reserve, a key indicator of heart disease risk. It showed strong performance in predicting cardiac outcomes. This may identify high-risk patients using a widely available and inexpensive 12-lead ECG, alleviating the low availability of costly PET scans.
No takes yet. Share an insight, caveat, or question.
Alahdab et al. (2024) conducted an observational in Coronary artery disease (n=21,288). Machine learning model using ECG to predict MFR vs. Baseline clinical models was evaluated on Major adverse cardiovascular events (MACEs) (HR 3.85 and 3.70). An ECG-based machine learning model predicting myocardial flow reserve <2 was associated with major cardiovascular events compared to baseline clinical models.
Synapse has enriched 10 related papers. Consider them for comparative context: