AI-assisted screening for heart valve disease using routine non-contrast chest CT

Artificial-Intelligence Assisted Opportunistic Screening for Valvular Heart Disease Using Non-contrast Chest CT Scans: A Prospective, Multicenter Study

This will test whether an AI program can spot moderate-to-severe heart valve disease on routine non-contrast chest CT scans in adults.

Quick facts

• Study type: Observational
• Enrollment: 3000 (estimated)
• Ages: 18 Years and up
• Sex: All
• Sponsor: Second Affiliated Hospital, School of Medicine, Zhejiang University (other)
• Drugs / interventions: radiation
• Locations: 3 sites (Wuhan, Hubei and 2 other locations)
• Trial ID: NCT07449130 on ClinicalTrials.gov

What this trial studies

This is a prospective, multicenter effort to validate a deep learning model that analyzes routine non-contrast chest CT images from adults seen in physical exams and outpatient or inpatient settings. The AI analyzes scans in real time and flags cases predicted to have moderate-to-severe valvular heart disease. Participants flagged by the AI will receive an immediate confirmatory echocardiogram, which serves as the diagnostic reference standard. The primary performance endpoint is sensitivity, with secondary endpoints including AUC, specificity, and accuracy, and results will be analyzed with 95% confidence intervals over an estimated 12-month enrollment period.

Who should consider this trial

Good fit: Adults (≥18) who had a routine non-contrast chest CT at a participating hospital and have complete medical records, including those flagged by the AI as having moderate-to-severe valve disease or selected negative controls.

Not a fit: Patients with poor-quality CT images, incomplete clinical records, implanted prosthetic valves, or scans obtained outside the qualifying timeframe are unlikely to benefit from this screening approach.

Why it matters

Potential benefit: If successful, this approach could catch moderate-to-severe valve disease earlier using CT scans patients already receive, enabling earlier follow-up and treatment.

How similar studies have performed: Some prior AI work has detected cardiac conditions from CT imaging, but large prospective multicenter validation specifically for valvular heart disease on non-contrast chest CT remains limited.

Eligibility criteria

Inclusion Criteria:

1. Age ≥ 18 years.
2. Complete electronic health record.
3. Non-contrast chest CT performed between Nov 1, 2025 - Nov 1, 2026 in any medical context (including physical exam, outpatient, inpatient, or emergency).
4. AI-predicted moderate or severe valvular heart disease, or deemed to require clinical intervention, or selected negative cases from sampling verification.

Exclusion Criteria:

1. Poor-quality non-contrast chest CT images.
2. Incomplete clinical records, involving severe deficiencies in critical diagnostic results, treatment records, imaging data, surgical records, medical history summaries, laboratory test results, or other essential medical information.
3. Presence of prosthetic valve implants, including aortic valves (mechanical valves, bioprosthetic valves), mitral valves (transcatheter edge-to-edge repair, bioprosthetic valves, mechanical valves, annuloplasty rings), tricuspid valves (TEER clipping, bioprosthetic valves, mechanical valves, annuloplasty rings), pulmonary valves (bioprosthetic valves), etc.
4. Abnormalities or conditions deemed by the investigator to warrant exclusion from the study enrollment.

Where this trial is running

Wuhan, Hubei and 2 other locations

• Renmin Hospital of Wuhan University — Wuhan, Hubei, China (RECRUITING)
• Xinjiang Uygur Autonomous Region People's Hospital — Ürümqi, Xinjiang, China (RECRUITING)
• The Second Affiliated Hospital of Zhejiang University School of Medicine — Hangzhou, Zhejiang, China (RECRUITING)

Study contacts

• Study coordinator: Jian'an Wang, MD

Conditions: Heart Valve Diseases