Artificial Intelligence to Scale Early Rheumatic Heart Disease Detection

NCT07599956 | Not Yet Recruiting

• 1 other identifier: 2026-0304
• Study Type: interventional
• Target: 62
• Locations: 1 country
• Sites: 1

Brief Summary

The main goal of this project is to see if RADAR (a machine-learning AI model) can help make rheumatic heart disease (RHD) screening easier to expand. Specifically, the project will test whether RADAR can screen as accurately-or more accurately-than current methods, and whether it can be used effectively in different low-resource settings. The aim is to show that RADAR could be adopted and used widely around the world.

Conditions

Rheumatic Heart Disease

Keywords

Rheumatic heart disease

Outcome Measures

Primary Outcomes (1)

• Accuracy of Provider RHD Screening

  The number of correctly identified (positive or negative) screenings divided by the total number of exams.

  1 year

Secondary Outcomes (1)

• Interpretation Sensitivity

  6 months

Study Arms (2)

Standard non-AI echocardiography

NO INTERVENTION

In the Standard non-AI Echocardiography arm, participants will receive the current standard of care under the ADUNU program, which includes a single parasternal long-axis view with black-and-white and color Doppler imaging. Providers have been trained to recognize mitral regurgitation greater than 1.5 or 2 cm (based on person's age38), any aortic insufficiency, qualitatively reduced left ventricular systolic function, and pericardial effusion. Detection of any of these findings constitutes a screen positive, prompting referral for a confirmatory echocardiogram.

RADAR-AI-assisted echocardiography

EXPERIMENTAL

In the RADAR-AI-Assisted Echocardiography arm, participants will undergo AI-assisted screening according to the well-established RADAR protocol including the same image acquisition protocol but interpreted by the tablet-based software based on two independent AI algorithms 1) RHD positive or negative and 2) mitral regurgitation jet length. Positive findings from either algorithm constitutes a screen positive. Providers may also refer for other concerns.

Diagnostic Test: AI assisted echocardiography

Interventions

AI assisted echocardiography DIAGNOSTIC_TEST

Continue standard of care with AI-assisted echocardiography

RADAR-AI-assisted echocardiography

Eligibility Criteria

• Sex: all
• Healthy Volunteers: Yes
• Age Groups: Child (0-17), Adult (18-64), Older Adult (65+)

You may qualify if:

• Employed at a participating ADUNU facility
• Holds a designated role in the ADUNU program as a nurse screener

You may not qualify if:

• None. The pragmatic trial design includes all eligible staff at participating facilities.

Sponsors & Collaborators

• Children's Hospital Medical Center, Cincinnati: lead
• Uganda Heart Institute: collaborator
• Children's National Health System, Washington DC: collaborator
• Ochsner Health System: collaborator

Study Sites (1)

Uganda Heart Institute

Kampala, Uganda

Location

MeSH Terms

Conditions

Rheumatic Heart Disease

Condition Hierarchy (Ancestors)

Rheumatic Fever; Streptococcal Infections; Gram-Positive Bacterial Infections; Bacterial Infections; Bacterial Infections and Mycoses; Infections; Heart Diseases; Cardiovascular Diseases

Study Officials

• Andrea Beaton

  Children's Hospital Medical Center, Cincinnati

  PRINCIPAL INVESTIGATOR

Central Study Contacts

Isabella Brigham

CONTACT

513-517-1307; isabella.aspromonte@cchmc.org

Study Design

• Study Type: interventional
• Phase: not applicable
• Allocation: RANDOMIZED
• Masking: SINGLE
• Who Masked: CARE PROVIDER
• Masking Details: Cardiologists who make up the adjudication panel.
• Purpose: DIAGNOSTIC
• Intervention Model: PARALLEL
• Sponsor Type: OTHER
• Responsible Party: SPONSOR

Model Details: Following informed consent, providers will be randomized using the computer-generated online randomization module in REDCap, housed at Cincinnati Children's. Randomization will use permuted blocks with no stratification. Of the 52 providers enrolled, 26 will be assigned to each study arm-AI-assisted echocardiography or standard non-AI echocardiography-in a 1:1 ratio.

Study Record Dates

• First Submitted: May 14, 2026
• First Posted: May 20, 2026
• Study Start (Estimated): June 1, 2026
• Primary Completion (Estimated): June 1, 2028
• Study Completion (Estimated): June 1, 2028
• Last Updated: May 20, 2026

Record last verified: 2026-05

Data Sharing

• IPD Sharing: Will not share

Locations

UG (1)