"SOUND" Trial: Study of On-site Use of Novel AI-assisted Diagnostics in CHD Screening

Efficacy Comparison Between Independent Auscultation by Primary Care Physicians and AI-Assisted Auscultation in Large-scale Screening for Congenital Heart Disease: A Cluster Randomized Controlled Trial in China

This study plans to conduct clinical validation of the model in real clinical settings, comparing it with primary care physicians and specialist physicians to ensure the model's practicality. Through continuous optimization and practice, the study aims to use AI-assisted heart sound auscultation to empower the auscultation capabilities of primary care obstetricians, pediatricians, and non-cardiovascular specialists nationwide. This will not only reduce the missed diagnosis rate and improve the detection rate of existing CHD screenings, but also expand the coverage of current CHD screening networks, incorporating newborns, infants, preschool children, children, and adolescents aged 0-18 years into the screening scope. The study aims to establish a new benchmark in child health management by providing feasible and cost-effective child health management solutions for other developing countries, contributing to global efforts for the health of children.

Study Overview

• Status: Completed
• Conditions: Congenital Heart Disease (CHD); Cluster Randomized Trial; Auscultation for Clinical Evaluation; Screening Tool; Artificial Intelligence (AI)
• Intervention / Treatment: Diagnostic test: Primary care physicians; Diagnostic test: AI + Primary care physicians

• Study Type: Interventional
• Enrollment (Actual): 19697
• Phase: Not Applicable

Contacts and Locations

Study Locations

• China
  • Qinghai
    • Xining, Qinghai, China, 810007
      • Qinghai Provincial Women and Children's Hospital

Participation Criteria

Eligibility Criteria

• Ages Eligible for Study: Child; Adult
• Accepts Healthy Volunteers: Yes

Description

Inclusion Criteria (Schools):

• The school is of the type: kindergarten, primary school, junior high school, or senior high school.
• The school has medical screening facilities and conditions that can support AI-assisted screening.
• The area where the school is located has at least one primary healthcare institution willing to participate in this trial.
• The school's management and teaching staff are willing to participate in the study and can cooperate to complete the related screening and data collection work.

Exclusion Criteria (Schools):

• More than half of the students in the school refuse to participate in the trial.
• Schools that are unable to complete the study due to severe limitations in geographical location and transportation conditions.
• Schools lacking medical screening facilities and conditions necessary for the implementation of screening.
• Areas where there are no primary healthcare institutions willing to participate in this trial.
• Schools whose management and teaching staff refuse to participate in the study or are unable to cooperate in completing the related screening and data collection work.

Inclusion Criteria (Individuals):

• Children aged between 0 and 18 years, regardless of gender.
• Children who agree to undergo echocardiography to determine the presence of congenital heart disease.
• Individuals who voluntarily participate in this study and sign the informed consent form.

Exclusion Criteria (Individuals):

• Individuals over the age of 18.
• Children who are unable to complete echocardiography or do not cooperate with auscultation.
• Participants who are unable to provide informed consent or are unwilling to comply with the study requirements to provide medical data for further analysis and research.
• Participants who have received interventional or surgical treatment for CHD.

Study Plan

How is the study designed?

Design Details

• Primary Purpose: Screening
• Allocation: Randomized
• Interventional Model: Parallel Assignment
• Masking: None (Open Label)

Number of Arms

2

Arms and Interventions

Participant Group / Arm	Intervention / Treatment
Active Comparator: Primary care physicians	Diagnostic test: Primary care physicians; For participants in Group A, a nonblinded independent staff member will first collect medical history, followed by sequential auscultation and CHD assessment by a specialist physician and a primary care physician, with an echocardiogram performed last.
Experimental: AI + Primary care physicians	Diagnostic test: AI + Primary care physicians; For participants in Group B, after medical history collection, both a specialist physician and a primary care physician will perform auscultation and CHD assessment. Subsequently, the primary care physician will use an electronic stethoscope to collect heart sound data according to the protocol and upload the recordings to a cloud platform. The AI model will analyze the data on the cloud platform and provide a diagnostic result within 5-10 seconds for the primary care physician's reference. The primary care physician may reassess the findings based on the AI model's feedback, and the participant will then undergo an echocardiogram.

What is the study measuring?

Primary Outcome Measures

Outcome Measure	Time Frame
Sensitivity of auscultation in identifying CHD between independent auscultation by primary care physicians and AI-assisted auscultation by primary care physicians	From enrollment to the end of treatment at 6 months

Secondary Outcome Measures

Outcome Measure	Time Frame
The rate of diagnostic revisions by physicians, the proportions of correct and incorrect changes	From enrollment to the end of treatment at 6 months
Sensitivity of Auscultation in CHD Detection: AI model & Experienced Cardiologists' Independent Auscultatio	From enrollment to the end of treatment at 6 months
Specificity of Auscultation in CHD Detection: Primary Care Physicians' Independent Auscultation & AI-assisted Primary Healthcare Physicians' Auscultation & AI model & Experienced Cardiologists' Independent Auscultation	From enrollment to the end of treatment at 6 months
Accuracy of Auscultation in CHD Detection: Primary Care Physicians' Independent Auscultation & AI-assisted Primary Care Physicians' Auscultation & AI model & Experienced Cardiologists' Independent Auscultation	From enrollment to the end of treatment at 6 months

Other Outcome Measures

Outcome Measure	Time Frame
Incidence rates of CHD in children across different regions of Qinghai Province	From enrollment to the end of treatment at 6 months

Collaborators and Investigators

• Sponsor: Kun Sun
• Collaborators: Bill and Melinda Gates Foundation

Study record dates

Study Major Dates

• Study Start (Actual): September 22, 2025
• Primary Completion (Actual): November 18, 2025
• Study Completion (Actual): January 4, 2026

Study Registration Dates

• First Submitted: January 18, 2025
• First Submitted That Met QC Criteria: January 18, 2025
• First Posted (Actual): January 24, 2025

Study Record Updates

• Last Update Posted (Actual): April 7, 2026
• Last Update Submitted That Met QC Criteria: April 1, 2026
• Last Verified: February 1, 2026

More Information

Terms related to this study

• Additional Relevant MeSH Terms: Cardiovascular Diseases; Heart Diseases; Congenital Abnormalities; Cardiovascular Abnormalities; Congenital, Hereditary, and Neonatal Diseases and Abnormalities; Heart Defects, Congenital
• Other Study ID Numbers: XHEC-C-2025-013-1; INV-072724 (Other Grant/Funding Number: Bill & Melinda Gates Foundation)

Plan for Individual participant data (IPD)

• Plan to Share Individual Participant Data (IPD)?: UNDECIDED
• IPD Plan Description: IPD may not be shared or may be unavailable for sharing due to several reasons. Firstly, confidentiality is of utmost importance. IPD contains sensitive personal information, and sharing such data could potentially compromise the privacy of participants, thereby restricting the sharing of such detailed information. Secondly, the data is protected by intellectual property rights. This may limit its sharing without proper agreements or permissions. Thirdly, IPD is large in volume and complex in nature, requiring substantial resources for storage, transmission, and analysis. Detailed discussions and plans need to be formulated to ensure the integrity and security of the data during the sharing process.

Drug and device information, study documents

• Studies a U.S. FDA-regulated drug product: No
• Studies a U.S. FDA-regulated device product: No
• product manufactured in and exported from the U.S.: No