Working…
COVID-19 is an emerging, rapidly evolving situation.
Get the latest public health information from CDC: https://www.coronavirus.gov.

Get the latest research information from NIH: https://www.nih.gov/coronavirus.
ClinicalTrials.gov
ClinicalTrials.gov Menu

Phono- and Electrocardiogram Assisted Detection of Valvular Disease (PEA-Valve)

The safety and scientific validity of this study is the responsibility of the study sponsor and investigators. Listing a study does not mean it has been evaluated by the U.S. Federal Government. Read our disclaimer for details.
 
ClinicalTrials.gov Identifier: NCT03458806
Recruitment Status : Active, not recruiting
First Posted : March 8, 2018
Last Update Posted : May 15, 2020
Sponsor:
Collaborator:
Eko Devices, Inc.
Information provided by (Responsible Party):
University of California, San Francisco

Brief Summary:

The diagnosis of valvular heart disease (VHD), or its absence, invariably requires cardiac imaging. A familiar and inexpensive tool to assist in the diagnosis or exclusion of significant VHD could both expedite access to life-saving therapies and reduce the need for costly testing. The FDA-approved Eko Duo device consists of a digital stethoscope and a single-lead electrocardiogram (ECG), which wirelessly pairs with the Eko Mobile application to allow for simultaneous recording and visualization of phono- and electrocardiograms. These features uniquely situate this device to accumulate large sets of auscultatory data on patients both with and without VHD.

In this study, the investigators seek to develop an automated system to identify VHD by phono- and electrocardiogram. Specifically, the investigators will attempt to develop machine learning algorithms to learn the phonocardiograms of patients with clinically important aortic stenosis (AS) or mitral regurgitation (MR), and then task the algorithms to identify subjects with clinically important VHD, as identified by a gold standard, from naïve phonocardiograms. The investigators anticipate that the study has the potential to revolutionize the diagnosis of VHD by providing a more accurate substitute to traditional auscultation.


Condition or disease Intervention/treatment
Aortic Valve Stenosis Mitral Regurgitation Heart Murmurs Valvular Heart Disease Diagnostic Test: AS Algorithm 1 Diagnostic Test: AS Algorithm 2 Diagnostic Test: MR Algorithm 1 Diagnostic Test: MR Algorithm 2

Show Show detailed description

Layout table for study information
Study Type : Observational
Estimated Enrollment : 900 participants
Observational Model: Case-Control
Time Perspective: Cross-Sectional
Official Title: Phono- and Electrocardiogram Assisted Detection of Valvular Disease
Actual Study Start Date : February 22, 2018
Estimated Primary Completion Date : December 31, 2020
Estimated Study Completion Date : December 31, 2020

Resource links provided by the National Library of Medicine

MedlinePlus related topics: Heart Diseases

Group/Cohort Intervention/treatment
Control
Subjects with echocardiographically confirmed valvular disease of less than moderate-to-severe grading with regards to aortic stenosis (AS) and mitral regurgitation (MR). Note that within this cohort will be a sub cohort consisting of subjects with structurally normal hearts, with no greater than mild valvular disease of any valve, no prior valvular intervention, and no evidence of congenital heart disease.
Diagnostic Test: AS Algorithm 2
Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater aortic stenosis from controls having any findings other than moderate-to-severe or greater aortic stenosis.

Diagnostic Test: MR Algorithm 2
Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater mitral regurgitation from controls having any findings other than moderate-to-severe or greater mitral regurgitation.

AS Case
Subjects with echocardiographically confirmed aortic stenosis (AS) of moderate-to-severe or greater grading.
Diagnostic Test: AS Algorithm 1
Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater aortic stenosis from controls having structurally normal hearts with no greater than mild valvular heart disease at any location.

Diagnostic Test: AS Algorithm 2
Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater aortic stenosis from controls having any findings other than moderate-to-severe or greater aortic stenosis.

MR Case
Subjects with echocardiographically confirmed mitral regurgitation (MR) of moderate-to-severe or greater grading.
Diagnostic Test: MR Algorithm 1
Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater mitral regurgitation from controls having structurally normal hearts with no greater than mild valvular heart disease at any location.

Diagnostic Test: MR Algorithm 2
Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater mitral regurgitation from controls having any findings other than moderate-to-severe or greater mitral regurgitation.

Control Subgroup
Subjects with structurally normal hearts, with no greater than mild valvular disease of any valve, no prior valvular intervention, and no evidence of congenital heart disease.
Diagnostic Test: AS Algorithm 1
Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater aortic stenosis from controls having structurally normal hearts with no greater than mild valvular heart disease at any location.

Diagnostic Test: MR Algorithm 1
Machine learning algorithm, generated from ECG and PCG recordings, distinguishing moderate-to-severe or greater mitral regurgitation from controls having structurally normal hearts with no greater than mild valvular heart disease at any location.




Primary Outcome Measures :
  1. Differentiation of clinically significant aortic stenosis from structurally normal hearts [ Time Frame: Close of study (after final enrollment of the aortic stenosis validation set), within 1 year. ]
    Identification by the trained machine learning algorithm of clinically important aortic stenosis (defined as moderate-to-severe or greater) from control subjects with structurally normal hearts and no greater than mild valvular heart disease, with comparison to the gold standard echocardiogram interpretation. As our algorithm will provide a continuous "score" to determine the likelihood of disease, the data will primarily come in the form of a receiver operating characteristic curve, for which we will calculate accuracy, specificity, and likelihood ratios at sensitivity cutoffs of 0.9, 0.95, and 0.99.

  2. Differentiation of clinically significant mitral stenosis from structurally normal hearts [ Time Frame: Close of study (after final enrollment of the mitral regurgitation validation set), within 1 year. ]
    Identification by the trained machine learning algorithm of clinically important mitral regurgitation (defined as moderate-to-severe or greater) from control subjects with structurally normal hearts and no greater than mild valvular heart disease, with comparison to the gold standard echocardiogram interpretation. As our algorithm will provide a continuous "score" to determine the likelihood of disease, the data will primarily come in the form of a receiver operating characteristic curve, for which we will calculate accuracy, specificity, and likelihood ratios at sensitivity cutoffs of 0.9, 0.95, and 0.99..


Secondary Outcome Measures :
  1. Differentiation of clinically significant aortic stenosis from the absence of clinically significant aortic stenosis [ Time Frame: Close of study (after final enrollment of the aortic stenosis validation set), within 1 year. ]
    Identification by the trained machine learning algorithm of clinically important aortic stenosis (defined as moderate-to-severe or greater) from controls with less than moderate-to-severe aortic stenosis, with comparison to the gold standard echocardiogram interpretation. As our algorithm will provide a continuous "score" to determine the likelihood of disease, the data will primarily come in the form of a receiver operating characteristic curve, for which we will calculate accuracy, specificity, and likelihood ratios at sensitivity cutoffs of 0.9, 0.95, and 0.99.

  2. Differentiation of clinically significant mitral regurgitation from the absence of clinically significant mitral regurgitation [ Time Frame: Close of study (after final enrollment of the mitral regurgitation validation set), within 1 year. ]
    Identification by the trained machine learning algorithm of clinically important mitral regurgitation (defined as moderate-to-severe or greater) from controls with less than moderate-to-severe mitral regurgitation, with comparison to the gold standard echocardiogram interpretation. As our algorithm will provide a continuous "score" to determine the likelihood of disease, the data will primarily come in the form of a receiver operating characteristic curve, for which we will calculate accuracy, specificity, and likelihood ratios at sensitivity cutoffs of 0.9, 0.95, and 0.99.



Information from the National Library of Medicine

Choosing to participate in a study is an important personal decision. Talk with your doctor and family members or friends about deciding to join a study. To learn more about this study, you or your doctor may contact the study research staff using the contacts provided below. For general information, Learn About Clinical Studies.


Layout table for eligibility information
Ages Eligible for Study:   18 Years and older   (Adult, Older Adult)
Sexes Eligible for Study:   All
Accepts Healthy Volunteers:   Yes
Sampling Method:   Non-Probability Sample
Study Population
Adults with either moderate-to-severe to severe AS or moderate-to-severe to severe MR (cases) and adults with structurally normal hearts with minimal VHD (controls). In practice, the accessible population will be adults meeting the entry criteria undergoing clinical echocardiograms at the UCSF echocardiography laboratory amenable to participation.
Criteria

Inclusion Criteria:

  • Able to provide consent
  • Undergoing a complete echocardiogram

Exclusion Criteria:

  • Refusal to participate

Information from the National Library of Medicine

To learn more about this study, you or your doctor may contact the study research staff using the contact information provided by the sponsor.

Please refer to this study by its ClinicalTrials.gov identifier (NCT number): NCT03458806


Locations
Layout table for location information
United States, California
University of California San Francisco
San Francisco, California, United States, 94143
Sponsors and Collaborators
University of California, San Francisco
Eko Devices, Inc.
Investigators
Layout table for investigator information
Principal Investigator: John Chorba, MD University of California, San Francisco
  Study Documents (Full-Text)

Documents provided by University of California, San Francisco:
Informed Consent Form  [PDF] March 1, 2018

Publications:

Layout table for additonal information
Responsible Party: University of California, San Francisco
ClinicalTrials.gov Identifier: NCT03458806    
Other Study ID Numbers: 17-21881
First Posted: March 8, 2018    Key Record Dates
Last Update Posted: May 15, 2020
Last Verified: May 2020
Individual Participant Data (IPD) Sharing Statement:
Plan to Share IPD: Undecided
Plan Description: We will create several de-identified databases of information and will be open to requests to share data as requested on a case-by-case basis.

Layout table for additional information
Studies a U.S. FDA-regulated Drug Product: No
Studies a U.S. FDA-regulated Device Product: Yes
Product Manufactured in and Exported from the U.S.: No
Keywords provided by University of California, San Francisco:
Auscultation
Phonocardiogram
Machine Learning
Heart Sounds
Additional relevant MeSH terms:
Layout table for MeSH terms
Heart Diseases
Mitral Valve Insufficiency
Aortic Valve Stenosis
Heart Valve Diseases
Heart Murmurs
Cardiovascular Diseases
Ventricular Outflow Obstruction
Signs and Symptoms