Screening for Cardiac and Cardiac-associated Pathology Using Single-channel Electrocardiogram

Official title

Screening for Cardiac and Cardiac-associated Pathology Using Single-channel Electrocardiogram Analyzed With Machine Learning Models

Overview

It is a prospective, controlled, single-center, observational, non-randomized study. The study is planned to include at least 4000 patients 18 years old and older in the training sample and 1000 patients over 18 years old in the test sample (the total number of patients is at least 5000 people). Patients will be included in the study if they have undergone a full examination (laboratory, clinical and instrumental), allowing for the verification or exclusion of cardiac and cardiac-associated pathology in accordance with current recommendations. During the course of the study, the authors of the work do not interfere with the above-mentioned scope of the examination, which is carried out on patients in accordance with clinical guidelines. All patients included in the study will undergo ECG recording in standard lead I for 1 minute twice, 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 will correlate with cardiac and cardiac-associated pathology

Detailed description

The aim of the study is to develop, evaluate the diagnostic efficacy, and pilot a screening method for cardiac and cardiac-related pathologies based on single-channel electrocardiogram analysis using artificial intelligence. To achieve this goal, a prospective, controlled, single-center, observational, non-randomized study will be conducted involving at least 4,000 patients meeting the inclusion criteria. All patients will undergo an expert echocardiography protocol with a comprehensive assessment of heart valve hemodynamics, systolic and diastolic myocardial function, along with standard measurements of the heart chambers and great vessels. To detect or exclude coronary artery disease: detection of significant coronary stenosis during coronary imaging: myocardial perfusion or determination of fractional coronary flow reserve, or stress echocardiography will be performed. For arterial hypertension: repeated blood pressure measurements in the office, 24-hour blood pressure monitoring will be performed. For cardiac arrhythmias: resting ECG and long-term Holter monitoring will be performed. For any cardiovascular pathology: blood tests: hemoglobin and red blood cell levels, glucose, glycated hemoglobin, oral glucose tolerance test, creatinine, uric acid, total cholesterol, low-density and high-density lipoproteins, and triglycerides will be performed. Additional testing data will be taken into account if conducted outside the protocol of this study. During the study, the authors will not interfere with the aforementioned scope of testing, which is performed on patients in accordance with clinical guidelines. All patients included in the study will undergo two ECG recordings 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 linking to the patients' personal data. A single-channel ECG will be recorded using a portable single-lead CardioQVARK ECG recorder. It was registered with the Federal Service for Surveillance in Healthcare on February 15, 2019. Number RU MI No. RZN 2024/22 855. 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. 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. The analysis involves the evaluation of the following parameters (the parameters listed below will be calculated as the median of the tact-cycle):• 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 - the total energy of the QRS wave based on the wavelet transform• T\_E - T-wave total energy based on wavelet transform• TP\_E- energy of the main tooth of the T-wave based on the wavelet transform• BETA, BETA\_S - T-wave asymmetry coefficients (simple and smooth 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 - the amplitude of the RF components of the QRS wave. Additionally used parameters:• TpTe, VAT, QTc - are duplicated to control the correctness of the record processing (the value of the UCC should be approximately equal to the median of the tick-by-bar).• QRSw - QRS width.• RA, SA, TA - the amplitudes of the R, S, T-waves, respectively, are used to normalize the parameters listed above.Statistical analysis and modeling will be performed using Python V3.8.8 and R V.4.0, as well as SPSS v.17.The correlation between various combinations of time, amplitude and frequency parameters of ECG and the presence of cardiac and cardiac-associated pathology will be analyzed. Certain parameters will be included in various multivariate analysis models: Lasso regression, Random Forest, Multilayer Perceptron, Support Vector Machine and Decision Tree. The model with the highest diagnostic accuracy will be selected, on which the algorithm will be tested.The outcome of this study will be the development and validation of an algorithm for identifying various cardiac and cardiac-associated pathologies based on the analysis of single-channel ECG parameters. The development of a medical-use program will also be undertaken.

Primary outcome measures

• Parameters of single-channel ECG that significantly correlate with the presence of various cardiac and cardiac-associated pathologies [Time frame: through study completion, an average of 2 years]
• Determination of sensitivity of various cardiac and cardiac-associated pathologies of multivariate models for analyzing single-channel electrocardiogram data [Time frame: through study completion, an average of 2 years]
• Determination of specificity of various cardiac and cardiac-associated pathologies of multivariate models for analyzing single-channel electrocardiogram data [Time frame: through study completion, an average of 2 years]
• Determination of diagnostic accuracy of various cardiac and cardiac-associated pathologies of multivariate models for analyzing single-channel electrocardiogram data [Time frame: through study completion, an average of 2 years]

Eligibility criteria

Inclusion criteria

• The presence of written informed consent of the patient to participate in the study
• Availability of examination data allowing for the verification or exclusion of cardiac and cardiac-associated pathology
• Age 18 years old and older

Non-inclusion criteria:

• Patients with an implanted permanent pacemaker;
• ECG changes that prevent spectral analysis;
• Conditions that may impair the quality of the ECG recording (Parkinson's disease, essential tremor, etc.);
• Conditions that make ECG recording in lead I impossible (congenital anomalies of the upper limbs, traumatic amputation of the upper limbs).
• Lack of written informed consent from the patient to participate in the study.

Exclusion criteria

• Poor quality of the ECG recording on a single-channel ECG monitor
• Insufficient examination data to verify or exclude cardiac or cardiac-associated pathology;
• Patient's unwillingness to continue participating in the study for any reason.

Criteria are shown verbatim from the registry (in English). Final eligibility is always assessed by the study center.

Healthy volunteers: Yes

Study design

Observational model

Cohort

Study locations

Center list to be confirmed — check the primary protocol.

Publications

• A screening method for predicting left ventricular dysfunction based on spectral analysis of a single-channel electrocardiogram using machine learning algorithms / N. Kuznetsova, Zh. Sagirova, A. Suvorov [et al.] // Biomedical Signal Processing and Control. - 2023. - Vol. 86. - P. 105219. - DOI 10.1016/j.bspc.2023.105219. - EDN APQSQF.
• Complex automated remote system for assessing hemodynamic parameters when analyzing the native signal of a single-channel ECG and pulse wave using machine learning techniques / N. O. Kuznetsova, Zh. N. Sagirova, E. A. Sultygova [et al.] // Russian Journal of Cardiology. - 2023. - T. 28, No. S7. - pp. 41-42. - EDN LZGDKG.
• Abdulraheem LW, Al-Dwa B, Shchekochikhin D, Gognieva D, Chomakhidze P, Kuznetsova N, Kopylov P, Bestavashvilli AA. A Systematic Review on the Effectiveness of Machine Learning in the Detection of Atrial Fibrillation. Curr Cardiol Rev. 2025;21(1):e310724232529. doi: 10.2174/011573403X293703240715104503. PMID 39092649
• Mesitskaya DF, Fashafsha ZZA, Poltavskaya MG, Andreev DA, Levshina AR, Sultygova EA, Gognieva D, Chomakhidze P, Kuznetsova N, Suvorov A, Marina I S, Poddubskaya E, Novikova A, Bykova A, Kopylov P. A single-lead ECG based cardiotoxicity detection in patients on polychemotherapy. Int J Cardiol Heart Vasc. 2024 Jan 20;50:101336. doi: 10.1016/j.ijcha.2024.101336. eCollection 2024 Feb. PMID 38304727
• Kuznetsova N.O., Alekseeva A.M., Mamedzade F.E., Sedov V.P., Kopylov F.Yu., Syrkin A.L., Chomakhidze P.Sh. Screening for heart defects when analyzing an electrocardiogram using machine learning methods (literature review) // Bulletin of new medical technologies. Electronic edition. 2025. No. 1. Publication 1-6. DOI: 10.24412/2075-4094-2025-1-1-6. EDN KNFSNS
• Kuznetsova N.O., Nartova A.A., Kurbanalieva N.K., Adueva D.Sh., Chursina E.Yu., Zhvania R.E., Ustinova D.I., Kostikova A.S., Kazakova M.V., Tarnaeva L.A., Chomakhidze P.Sh., Kopylov F.Yu. Results of screening for heart rhythm disturbances using a single-channel electrocardiogram without the participation of medical personnel // Bulletin of new medical technologies. 2025. No. 1. P. 56-60. DOI: 10.2

Identifiers

NCT: NCT07396792 · 7747-211021

Primary sources (government registries)

• ClinicalTrials.gov (US NIH)

View this study on ClinicalTrials.gov ↗