AI-Enhanced Single-Lead ECG Screening for Coronary Stenosis

Overview

It is a prospective, controlled, single-center, non-randomized, observational study. Two patient groups are planned for inclusion: the first - 200 patients with significant coronary artery stenosis confirmed by coronary angiography (CAG) or multislice computed tomography (MSCT) results; the second - a control group consisting of 200 patients without significant stenosis according to CAG or MSCT data. All study subjects will have a date of coronary artery imaging via CAG or MSCT with assessment of myocardial perfusion. Stress echocardiography tests or fractional flow reserve (FFR) assessment will be conducted as indicated. All patients included in the study will undergo ECG recording within 1 month before or after CAG or MSCT in standard lead I for 1 minute, followed by spectral analysis of the obtained data, which will be stored at the remote monitoring center of Sechenov University without being linked to the personal data of patients. A spectral analysis of the electrocardiogram will be performed using a continuous wavelet transform. The result of this study will be the identification of ECG parameters that correlate with significant coronary artery stenosis.

The aim of the study:: To develop and evaluate the diagnostic efficacy of a screening method for significant coronary artery stenosis based on data obtained from the analysis of a single-channel electrocardiogram. This is a prospective, controlled, single-center, non-randomized, observational study. Two patient groups are planned for inclusion: the first group comprises 200 patients with significant coronary artery stenosis confirmed by coronary angiography (CAG) or multislice computed tomography (MSCT) results; the second group is a control group consisting of 200 patients without significant stenosis according to CAG or MSCT data. All study subjects will have a date of coronary artery imaging via CAG or MSCT with assessment of myocardial perfusion. Stress echocardiography tests or fractional flow reserve (FFR) assessment will be conducted as clinically indicated. ECG registration in standard lead I will be performed within 3 months before or after the CAG or MSCT. Obtained data will be stored at the remote monitoring center of Sechenov University without being linked to the personal data of patients. A spectral analysis of the electrocardiogram will be performed using a continuous wavelet transform. The single-channel ECG will be recorded using the portable single-lead ECG monitor CardioQvark. It is designed as an iPhone cover. It is registered with the Federal Service for Health Surveillance on February 15, 2019. RZN No. 2019/8124. The result of this study will be the identification of ECG parameters that correlate with significant coronary artery stenosis. The patient's personal data (last name, first name, patronymic, date of birth, contact information) will not be transferred or taken into account. Each patient is assigned an individual number that is not associated with his/her personal data. Subsequently, spectral analysis of the electrocardiogram will be performed using machine learning models and/or neural network data analysis. Then a spectral analysis of the electrocardiogram will be performed using a continuous wavelet transform, the principles of which are based on the Fourier transform. Analysis of the single-channel ECG involves evaluation of the following parameters (the parameters listed below will be calculated as median beat-to-beat values): * TpTe - time from peak to end of the T-wave * VAT - time from the beginning of the QRS to the R-peak * QTc - corrected QT interval. * QT/TQ - the ratio of QT length to TQ length (from the end of T to the beginning of the QRS of the next complex). * QRS\_E - total energy of the QRS-wave based on wavelet transform * T\_E - total energy of the T-wave based on wavelet transform * TP\_E - energy of the main T-wave peak based on wavelet transform * BETA, BETA\_S - T-wave asymmetry coefficients (simple and smoothed versions) * BAD\_T - flag of T-wave quality (whether expressed in the current lead) * QRS\_D1\_ons - energy of the leading edge of the R-wave (based on the "first derivative" wavelet transform) * QRS\_D1\_offs - energy of the trailing edge of the R-wave (based on the "first derivative" wavelet transform) * QRS\_D2 - peak energy of the R-wave (based on the "second derivative" wavelet transform) * QRS\_Ei (i=1,2,3,4) - QRS-wave energy in 4 frequency ranges (2-4-8-16-32 Hz) based on wavelet transform * T\_Ei (i=1,2,3,4) - T-wave energy in 4 frequency ranges (2-4-6-8-10 Hz) based on wavelet transform * HFQRS - amplitude of the high-frequency components of the QRS-wave Additionally used parameters: * TpTe, VAT, QTc - are duplicated to control the correctness of record processing (the value of the central measure should be approximately equal to the beat-to-beat median). * QRSw - QRS width. * RA, SA, TA - amplitudes of the R, S, T-waves, respectively, used for normalizing the parameters listed above. Method of statistical processing of results: SPSS Statistics Version 26 computer program for statistical data processing; construction of machine learning models and/or neural network data analysis The proposed research outcome: development of an algorithm for diagnosing significant coronary stenosis based on single-channel ECG data using elements of artificial intelligence. The endpoints of the study are the parameters of diagnostic accuracy of the developed model: * specificity, * sensitivity, * prognostic significance of a positive and negative result, * diagnostic accuracy. Тhese metrics will be calculated using receiver operating characteristic (ROC) analysis and confusion matrices on a held-out test set (30% of the dataset) after training multifactorial models (logistic regression, random forest, or neural networks) on single-lead ECG features. Sensitivity, specificity, positive/negative predictive values, and overall accuracy will be derived by comparing model predictions of significant coronary stenosis (≥50% lumen narrowing per CAG/MSCT) against the gold standard, with cross-validation (k=5 folds) to ensure robustness and bootstrap resampling for 95% confidence intervals.