Project Summary/Abstract Tuberculosis (TB) remains a global public health problem. Globally, 10 million people develop active TB each year, but one-third of them are not diagnosed or started on treatment. Systematic screening of high-risk populations, known as “active case finding,” can facilitate early diagnosis and reduce the global TB burden. While chest X-ray (CXR) is a sensitive tool for TB screening, high-burden countries often do not have enough qualified readers needed to scale up CXR-based screening. Advances in artificial intelligence (AI) offer a promising alternative through computer-aided detection (CAD). CAD systems analyze a CXR for signs of TB and generate a numeric score that can be used to select people for further testing. Recently endorsed by the WHO for TB screening, commercial CAD systems have begun their deployment for active case finding. However, for CAD to realize its full potential and have a meaningful impact on TB epidemiology, it is essential to tailor its implementation to the local population and screening context. This proposal aims to optimize the implementation of digital X-ray technology with CAD for community-based TB screening in sub-Saharan Africa. This will be accomplished by evaluating two novel screening strategies against the current standard approach, which is to offer sputum testing to individuals with an X-ray abnormality score above a set threshold. The first strategy is to individually adjust this threshold according to client characteristics such as age, sex, HIV status, and symptoms (Aim 1). The second strategy is to develop and utilize an independent CAD model, trained on chest radiographs from the local screening population, as opposed to relying on a commercial CAD product trained on a larger, but less representative dataset (Aim 2). In addition to evaluating the diagnostic accuracy of these approaches, the feasibility and cost-effectiveness of each strategy will be evaluated against the conventional method within the Ugandan context (Aim 3). This mentored patient-oriented research will not only inform future implementation of CAD for TB screening but also provide a robust training platform for the award recipient. Through both research and career development training, the recipient will acquire essential skills in advanced statistics, AI analytics, implementation science, and health economics, as well as hands-on experience in field data collection. This will lay the foundation for an independent career as a clinical investigator focused on the implementation of AI-driven health innovations in resource-limited settings.