Artificial Intelligence (AI) Analysis of Synchronized Phonocardiography (PCG) and Electrocardiogram(ECG)

A Deep-learning-based Multi-modal Phonocardiogram(PCG) and Electrocardiogram(ECG) Processing Framework for Screening Depressed Left Ventricular Ejection Fraction (dLVEF) Using a Wearable Cardiac Patch

Status

Recruiting

Phases

Unknown

Study type

Observational

Source

ClinicalTrials.gov

Registry ID

NCT06009718

Enrollment

3000

Registered

2023-08-24

Start date

2023-08-25

Completion date

2028-06-01

Last updated

2025-01-16

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Heart Failure

Keywords

Phonocardiography (PCG), Electrocardiogram (ECG), Heart Failure (HF), depressed left ventricular ejection fraction (dLVEF)

Brief summary

The diagnosis of depressed left ventricular ejection fraction (dLVEF) (EF\<50%) depends on golden standard ultrasound cardiography (UCG). A wearable synchronized phonocardiography (PCG) and electrocardiogram (ECG) device can assist in the diagnosis of dLVEF, which can both expedite access to life-saving therapies and reduce the need for costly testing.

Detailed description

The synchronized PCG and ECG is wirelessly paired with the WenXin Mobile application, allowing for simultaneous recording and visualization of PCG and ECG. These features uniquely enable this device to accumulate large sets of acoustic data on patients both with and without heart failure(HF). This study is a Case-control study. In this study, the investigators seek to develop an artificial intelligence (AI) analysis system to identify dLVEF (EF\<50%) by PCG and ECG. All adults (aged ≥18 years) planned for UCG were eligible to participate (inpatients and outpatients). Specifically, the investigators will attempt to develop machine learning algorithms to learn synchronized PCG and ECG of patients with dLVEF. Then we use these algorithms to identify dLVEF subjects. The investigators anticipate to demonstrate the wearable cardiac patch with synchronized PCG and ECG can reliably and accurately diagnose dLVEF in the primary care setting.

Interventions

None listed

Study design

Observational model

CASE_CONTROL

Time perspective

PROSPECTIVE

Eligibility

Sex/Gender

ALL

Age

18 Years to 100 Years

Healthy volunteers

Inclusion criteria

* Attendance at RuiJin hospital for UCG * Signed dated informed consent * Commit to follow the research procedures and cooperate in the implementation of the whole process research * UCG has been completed * Age ≥ 18 * At least 8 consecutive cycles of sinus rhythm can be recorded

Exclusion criteria

* Patients with pacemakers * Complete left bundle branch block or block or QRS wave widening\>120ms * Left chest skin damaged or allergic to patch * Refusal to participate

Design outcomes

Primary

Measure	Time frame	Description
Determination of Heart Failure Disease	one time assessment at baseline (approx. 5 minutes)	Heart Failure Disease was determined by EMAT （millisecond, ms）calculate from synchronized PCG and ECG signals using an artificial intelligence (AI) guided model.

Countries

China

Contacts

Primary ContactWenli Zhang, MD

zwl11929@rjh.com.cn+86 21 13917615339

Backup ContactBei Song, MD

belasong@163.com+86 21 15821960139

Outcome results