Using Artificial Intelligence to Improve Tuberculosis Screening with Chest X-rays in Primary Healthcare Settings in China

What this Trial Is About

Tuberculosis TB remains a major global health challenge with high rates of illness and death, making early and accessible detection critical. This research evaluates the impact of an artificial intelligence AI-assisted chest X-ray system JF CXR-1 in improving TB diagnosis speed and accuracy in primary care settings in Yichang City, China, which faces a high TB burden. The study aims to assess how integrating AI into the diagnosis process can enhance the overall TB care pathway and support public health goals. The study is conducted in three stages first, retrospective imaging will be used to optimize and validate the AI-assisted systems accuracy. Second, a cluster randomized controlled trial will compare the AI-assisted chest X-ray screening to routine doctor analysis in township medical institutions, measuring diagnostic times and accuracy. Third, qualitative and quantitative methods will evaluate the feasibility and practicality of applying the AI screening approach in various community environments, aiming for wider adoption in primary healthcare. Participants aged over 15 years with TB-related respiratory symptoms who have not been previously diagnosed with active pulmonary TB will undergo chest X-rays. Those in the intervention group will have their X-rays analyzed by the AI-assisted system alongside doctors, while the control group will have standard doctor-only reviews. Researchers will collect baseline data, follow diagnosis timing through the TB disease system, and assess diagnostic yield differences over six and twelve months. The study includes follow-up and monitoring to evaluate the AI systems impact on TB diagnosis and care efficiency.

Effectiveness of Computer-Aided Detection Chest X-Ray Screening for Improving Tuberculosis Diagnostic Yield in Chinese Primary Health Care Settings: Study Protocol for a Prospective Cluster Randomized Controlled Trial