Back

Accelerating Tuberculosis Diagnosis in Mozambican Prisons Using Digital Chest X-rays with Computer Aided Detection: Preliminary Results from a Longitudinal, Comprehensive Health Intervention.

Olotu, A. A.; Chiramal, J. A.; Boehm, R. A.; Nair, A. M.; Chawla, S.; Vicente, M. A.; Matusse, D. A.; Uate, S. T.; Munguambe, A. S.; Prophete, E. J.; Brown, V. M.; Anli, C. M.; Vijayan, S.; Calder, I. R.; Spaulding, A. C.

2024-12-02 public and global health
10.1101/2024.12.01.24318070 medRxiv
Show abstract

BackgroundMozambique has a high burden of tuberculosis (TB) and in 2021, an estimated 18,000 incident cases nationwide were not diagnosed. Global estimates suggest that 47% of TB disease in prisons is undiagnosed. We implemented an integrated-care-model health intervention to enhance the diagnosis and treatment of TB disease, provide TB preventive treatment (TPT), and identify and treat other undiagnosed health conditions. MethodsBeginning from July 11, 2023, and through the 2nd quarter of 2024 we conducted active case-finding for TB using digital chest x-rays with computer aided detection (DCXR-CAD), in three prisons in Maputo, Mozambique. We conducted clinical assessments for other health conditions and delivered TPT. ResultsOf 7912 individuals screened, 264 new cases of TB disease were notified, a TB screening yield of 3.34% and a number needed to screen of 30, and 1346 persons were initiated on TPT. Other conditions were diagnosed and treated including HIV (28), malnutrition (830), and skin conditions (462). ConclusionsStrengthening local infrastructure and implementing DCXR-CAD for screening gave a substantial TB yield in this population. Paying attention simultaneously to preventing TB and addressing other health conditions in this vulnerable population was important.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.