MARATTO

preprint · medRxiv

Implementation of digital chest radiography for childhood tuberculosis diagnosis at district hospital level in six high tuberculosis burden and resources limited countries

2024Open accessUniversity of Douala

Abstract

Abstract Objectives Chest X-ray (CXR) plays an important role in childhood tuberculosis (TB) diagnosis but access to quality CXR remains a major challenge in resource-limited settings. Digital CXR (d-CXR) can solve some image quality issues and facilitate their transfer for quality control. We describe the implementation of introducing d-CXR in twelve district hospitals (DH) in 2021-22 across Cambodia, Cameroon, Ivory Coast, Mozambique, Sierra Leone and Uganda as part of the TB-Speed decentralization study on childhood tuberculosis diagnosis. Methods For digitization of CXR Digital Radiography (DR) plates was set-up on existing analogue radiography machines. D-CXR were transferred to an international server at Bordeaux University and downloaded by sites’ clinicians for interpretation. We assessed the pre-intervention (baseline situation and d-CXR set-up) and per-intervention (uptake, challenges and health care workers’ (HCW) perceptions) of d-CXR implementation. We used a convergent mixed method approach utilizing process data, individual interviews with 113 HCWs involved in performing or interpreting d-CXRs and site support supervision reports. Results Of 3104 children with presumptive TB, 1642 (52.9%) had at least one d-CXR including 1505, 136 and 1 children with one, two and three d-CXR respectively, resulting in a total of 1780 d-CXR. Of them, 1773 (99.6%) were of good quality and 1772/1773 (99.9%) were interpreted by sites’ clinicians. 164 children had no d-CXR performed despite attending the radiography department: 126, 37 and 1 with one, two and three attempts, respectively. D-CXRs were not performed in 21.6% (44/203) due to connectivity problem between the DR plate captor and the computer. HCWs reported good perceptions of d-CXR and of the DR plates provided. The main challenge was the upload to and download from the server of d-CXRs, due to limited internet access. Conclusion D-CXR using DR plates was feasible at district hospital level and provided good quality images but required overcoming operational challenges.

Research topics

  • COVID-19 diagnosis using AI
  • Lung Cancer Diagnosis and Treatment
  • Radiomics and Machine Learning in Medical Imaging

Read the original research

This page summarises published work. The authoritative version sits with the publisher.

DOI: 10.1101/2024.08.23.24312489

Is something wrong with this record? Report it or request removal.

Discussion

Discuss this research

Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.

No discussion yet. Open the first thread.