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

Not Applicable

Not yet recruiting

• Conditions: Congenital Heart Disease (CHD); Artificial Intelligence (AI); Cluster Randomized Trial; Screening Tool; Auscultation for Clinical Evaluation

• Registration Number: NCT06791109
• Lead Sponsor: Kun Sun

Brief Summary

In recent years, the integration of artificial intelligence (AI) into healthcare has accelerated, with deep learning emerging as a powerful tool in medicine. Deep learning algorithms can automatically extract complex features and patterns, enabling highly accurate recognition of heart sounds. Building on this foundation, Professor Sun Kun and his team at Xinhua Hospital, affiliated with Shanghai Jiao Tong University School of Medicine, in collaboration with multiple centers across China, have been investigating the development and application of AI-assisted heart sound recognition for screening congenital heart disease (CHD). By leveraging electronic stethoscopes to collect heart sounds and applying AI algorithms to extensive datasets from both children with CHD and healthy children, the system has been trained to discern between normal and pathological murmurs, achieving an accuracy and sensitivity rate of 90%.

This prospective cluster-randomized clinical trial is designed to assess the non-inferiority of AI-assisted primary healthcare physicians in identifying CHD compared to specialist physicians conducting independent identification. Through ongoing optimization and broad application, this research endeavors to equip general practitioners, child health physicians, and non-cardiovascular specialists with AI-assisted electronic auscultation capabilities, ultimately enhancing the standard of pediatric care nationwide. As the screening network expands, the project is expected to reduce missed diagnoses and increase detection rates in CHD screening. Moreover, the project aspires to offer feasible and cost-effective child health management solutions to other developing countries, contributing to global efforts to improve children's health.

Detailed Description

Not available

Study Timeline

• Status Verified: 2025-01
• Study First Submitted: 2025-01-18
• Study First Submitted QC: 2025-01-18
• Last Update Submitted: 2025-01-18
• Study First Posted: 2025-01-24
• Last Update Posted: 2025-01-24
• Study Start: 2025-03-01
• Primary Completion: 2025-09-01
• Study Completion: 2025-12-31

Recruitment & Eligibility

• Status: NOT_YET_RECRUITING
• Sex: All
• Target Recruitment: 70590

Inclusion Criteria

Not provided

Exclusion Criteria

Not provided

Study & Design

• Study Type: INTERVENTIONAL
• Study Design: PARALLEL

Primary Outcome Measures

Name	Time	Method
Sensitivity of Auscultation in CHD Detection: Experienced Cardiologists' Independent Auscultation & AI-assisted Primary Healthcare Physicians' Auscultation	From enrollment to the end of treatment at 6 months

Secondary Outcome Measures

Name	Time	Method
Specificity, accuracy, and false negatives of Auscultation in CHD Detection: Experienced Cardiologists' Independent Auscultation & AI-assisted Primary Healthcare Physicians' Auscultation	From enrollment to the end of treatment at 6 months
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, specificity, accuracy, and false negatives of auscultation in CHD detection: Primary Care Physicians/Experienced Cardiologists' Independent Auscultation & AI Model	From enrollment to the end of treatment at 6 months