Comparative effectiveness of chest ultrasound, chest X-ray and computer-aided diagnostic (CAD) for tuberculosis diagnosis in low-resource setting: study protocol for a cross-sectional study from Ethiopia

Introduction: Early and accurate diagnosis of pulmonary tuberculosis (TB) is crucial for timely treatment and prevention of transmission, but diagnostic challenges persist due to complex symptoms and limitations in diagnostic tools. Chest X-ray (CXR) is the standard imaging modality, but its sensitivity and specificity may vary. Recently, some promising alternatives emerged such as chest ultrasonography (CUS) – particularly valuable in resource-limited settings – and computer-aided diagnosis (CAD) systems – helping clinicians in the reading and interpretation of the CXR. However, direct comparisons of CUS, CXR, and CAD score in TB diagnosis are limited.

Methods and analysis: This cross-sectional study will assess the diagnostic effectiveness of CUS in diagnosing TB compared to CXR and CAD score among index cases and household contacts. The study will be conducted at Wolisso St. Luke Hospital (Wolisso, Ethiopia). Index cases will be subjects with diagnosis of pulmonary tuberculosis within 7 days. Household contacts will be identified by administering a screening questionnaire to index cases. They will undergo CXR as for standard of care and consequent CAD analysis and CUS. The anticipated sample size is 136 subjects. The common accuracy metrics (sensitivity, specificity, positive and negative predictive values) will be calculated.

Ethics and dissemination: The protocol was approved by the Oromia Health Bureau Research Ethics Committee (BFO/MBTFH/1-16/1908). All information obtained will be confidential. Selected investigators will have access to data, while international partners will sign a dedicated Data Protection Agreement. Eligible subjects will receive a brief information about the study before being asked to participate and they will provide a written informed consent. Results will be conveyed to stakeholders and disseminated through conferences and peer-reviewed journals.

Clinical trial registration: NCT06409780, https://clinicaltrials.gov.