Using an AI-assisted chest X-ray to find tuberculosis faster in primary care in Yichang, China

Artificial Intelligence-assisted Chest X-ray in TB Screening: Effectiveness of Enhancing the Care Cascades in Chinese Primary-care Settings (ACCESS-CARE)

This project will test whether adding a certified AI tool to chest X-rays helps find active pulmonary TB faster and more accurately for people over 15 with TB symptoms who visit township primary care clinics in Yichang.

Quick facts

• Phase: NA
• Study type: Interventional
• Enrollment: 22000 (estimated)
• Ages: 15 Years and up
• Sex: All
• Sponsor: Peking Union Medical College (other)
• Locations: 1 site (Yichang, Hubei)
• Trial ID: NCT06963606 on ClinicalTrials.gov

What this trial studies

This is a cluster randomized controlled trial randomizing township primary healthcare centers in Yichang City to use a certified AI-assisted chest X-ray system (JF CXR-1) plus routine doctor review versus routine chest X-ray reading alone. Eligible participants over 15 with TB-related respiratory symptoms will complete a baseline questionnaire, receive chest radiography, and undergo pathogen testing and follow-up as indicated. A 100-person pre-trial was conducted to refine procedures before the full cluster trial. Primary outcomes include diagnostic yield for pulmonary TB and time to diagnosis, with secondary outcomes examining linkage to testing and treatment along the TB care cascade.

Who should consider this trial

Good fit: Ideal candidates are people aged over 15 who visit township primary care in Yichang with TB-related respiratory symptoms, have not been previously diagnosed with active pulmonary TB, and can complete pathogen testing and follow-up.

Not a fit: People with extrapulmonary or latent TB, those under 15, individuals with poor-quality chest X-rays, or those unable to complete follow-up or pathogen testing are unlikely to benefit from this screening approach.

Why it matters

Potential benefit: If successful, this approach could speed up diagnosis and get symptomatic patients into testing and treatment sooner, which may reduce transmission.

How similar studies have performed: Automated chest X-ray interpretation tools have shown promise in prior diagnostic and screening studies and have received WHO guidance for certain screening uses, but their effect on the entire care cascade and population-level diagnostic time remains less well established.

Eligibility criteria

Eligibility criteria Inclusion criteria

Participants must receive medical treatment at the primary healthcare hospitals in Yichang City, Hubei Province, and underwent chest X-ray examinations. The participants have to meet the following criteria:

1. \>15 years old.
2. Appearance of tuberculosis-related respiratory symptoms or signs.
3. Individuals not previous diagnosed with active pulmonary tuberculosis.
4. Capable of completing pathogen examinations and subsequent related inspections.

Exclusion criteria

Those who meet any of the below criteria will be excluded:

1. Diagnosed with extrapulmonary tuberculosis or latent tuberculosis infection during the current visit.
2. The quality of Chest X-ray images did not meet the standard requirements.
3. Unrecognized identity information participants.

Withdrawal Criteria

1. Participants who are lost to follow-up or who do not complete the follow-up period.
2. Participants who experience a sudden and serious illness or choose not to continue participating in the study.

Where this trial is running

Yichang, Hubei

• Township Health-care settings in Yichang City — Yichang, Hubei, China (RECRUITING)

Study contacts

• Study coordinator: Yang Xuelin

Conditions: Tuberculosis, tuberculosis, computer-assisted detection