Screening for valvular heart disease with a single-lead ECG

Screening of Valvular Heart Disease Using Single-channel Electrocardiogram Analyzed With Machine Learning Models

This project will test whether a one-minute single-lead ECG can detect valve problems in adults by comparing ECG recordings to echocardiogram results.

Quick facts

• Study type: Observational
• Enrollment: 1200 (estimated)
• Ages: 18 Years and up
• Sex: All
• Sponsor: I.M. Sechenov First Moscow State Medical University (other)
• Locations: 1 site (Moscow)
• Trial ID: NCT07099417 on ClinicalTrials.gov

What this trial studies

This is a prospective, controlled, single-center observational study enrolling at least 1,200 adults at Sechenov University. Each participant will receive a standard echocardiogram read independently by two experts, followed immediately by a one-minute single-lead (lead I) ECG recorded with a CardioQvark device. ECG signals will be deidentified and stored remotely, then analyzed using spectral methods including continuous wavelet transform and machine-learning model development with a 1,000-patient training set and 200-patient test set. The aim is to identify ECG parameters and build models that correlate with echocardiographic findings of valvular stenosis or regurgitation.

Who should consider this trial

Good fit: Adults aged 18 and older who can give written informed consent and can attend outpatient or inpatient evaluation at the Sechenov University center are eligible, provided they do not have pacemakers or prosthetic valves and can produce an adequate ECG and echocardiogram.

Not a fit: Patients with pacemakers, prosthetic valves, movement disorders that impair ECG quality, poor echocardiographic windows, or acute illnesses that prevent completion of tests are unlikely to benefit from this screening approach.

Why it matters

Potential benefit: If successful, this could enable a fast, low-cost screening tool to flag patients who need further echocardiographic evaluation for valve disease.

How similar studies have performed: Using single-lead ECGs and machine learning to detect structural heart problems is relatively novel: there are preliminary pilot studies suggesting signal-based detection is possible, but large-scale validation for valvular disease is limited.

Eligibility criteria

Inclusion Criteria:

* The presence of written informed consent of the patient to participate in the study
* Age from 18 years
* Outpatient treatment and / or hospitalization in a research center

Exclusion Criteria:

* Reluctance of the patient to participate in the study
* Poor quality ECG recording on a single-channel ECG monitor
* Poor visualization of the heart during echocardiographic study
* Acute psychotic reactions that arose during research
* An exacerbation of chronic diseases requiring treatment tactics for the patient and preventing his further participation in the study.

Non-inclusion criteria:

* Poor quality ECG recording on a single-channel ECG monitor
* Conditions that can impair ECG recording quality (Parkinson's disease, essential tremor)
* Mental illness
* Patients with a pacemaker installed
* Patients with prosthetic valves

Where this trial is running

Moscow

• I.M. Sechenov First Moscow State Medical University (Sechenov University) — Moscow, Russia (RECRUITING)

Study contacts

• Principal investigator: Natalia Kuznetsova, Dr. — I.M. Sechenov First Moscow State Medical University (Sechenov University)
• Study coordinator: Natalia Kuznetsova, Dr.

Conditions: Valvular Heart Disease Stenosis and Regurgitation, Valvular Heart Disease, valvular heart disease, Screening, single-channel electrocardiogram, machine learning models