South Africa, Zambia and Zimbabwe – Finding asymptomatic TB in high-risk groups with Delft Ultra & CAD4TB
Community-based Active Case Finding (ACF) is crucial for identifying undiagnosed TB cases, including those without symptoms who may contribute to community-based transmission. The XACT-19 project incorporates ultra-portable X-rays with CAD/AI to enhance TB detection in the community.
Multi-country study deployment
The study was conducted in South Africa, Zambia, and Zimbabwe using the ultra-portable Delft Ultra X-ray with CAD4TB version 7.
Targeting high-risk populations
Screening focused on symptomatic individuals, people living with HIV, those with previous TB, people with diabetes, and TB contacts.
Comparative study design
Participants were assigned to two study arms: CAD4TB with Xpert and Xpert-only screening.
The interim results showed
- Among the Bac+ cases, 20% of asymptomatic individuals were found to be infectious.
- The study highlighted an instance of a relatively healthy, asymptomatic individual who was TB-positive, suggesting this person could be unknowingly transmitting the disease in the community.
- The XACT-19 project demonstrated the effectiveness of using Delft Ultra with CAD4TB in detecting high TB burden settings, suggesting a significant minority (~20%) was probably infectious, with a large proportion asymptomatic.
- These findings inform future ACF strategies in high TB burden settings, which is key for implementing strategies to detect TB in the community using these tools.
REFERENCE: Scott, A. et al (2024, November 12-16). Can AI-driven computer-aided detection optimize X-pert-oriented community-based active case finding for TB? An interim trial progress report [Presentation]. The Union World Conference on Lung Health, Bali, Indonesia.