Effectiveness and cost-effectiveness of community-based TB screening algorithms using computer-aided detection (CAD) technology alone compared with CAD combined with point-of-care C reactive protein testing in Lesotho and South Africa: protocol for a paired screen-positive trial
Bibliographic record
Abstract
Introduction Tuberculosis (TB) remains a significant public health challenge in many African communities, where underreporting and underdiagnosis are prevalent due to barriers in accessing care and inadequate diagnostic tools. This is particularly concerning in hard-to-reach areas with a high burden of TB/HIV co-infection, where missed or delayed diagnoses exacerbate disease transmission, increase mortality and lead to severe economic and health consequences. To address these challenges, it is crucial to evaluate innovative, cost-effective, community-based screening strategies that can improve early detection and linkage to care. Methods and analysis We conduct a prospective, community-based, diagnostic, pragmatic trial in communities of the Butha Buthe District in Lesotho and the Greater Edendale area of Msunduzi Municipality, KwaZulu-Natal in South Africa to compare two strategies for population-based TB screening: computer-aided detection (CAD) technology alone (CAD4TBv7 approach) versus CAD combined with point-of-care C reactive protein (CRP) testing (CAD4TBv7-CRP approach). Following a chest X-ray, CAD produces an abnormality score, which indicates the likelihood of TB. Score thresholds informing the screening logic for both approaches were determined based on the WHO’s target product profile for a TB screening test. CAD scores above a threshold prespecified for the CAD4TBv7 approach indicate confirmatory testing for TB (Xpert MTB/RIF Ultra). For the CAD4TBv7-CRP approach, a CAD score within a predefined window requires the conduct of the second screening test, CRP, while a score above the respective upper threshold is followed by Xpert MTB/RIF Ultra. A CRP result above the selected cut-off also requires a confirmatory TB test. Participants with CAD scores below the (lower) threshold and those with CRP levels below the cut-off are considered screen-negative. The trial aims to compare the yield of detected TB cases and cost-effectiveness between two screening approaches by applying a paired screen-positive design. 20 000 adult participants will be enrolled and will receive a posterior anterior digital chest X-ray which is analysed by CAD software. Ethics and dissemination The protocol was approved by National Health Research Ethics Committee in Lesotho (NH-REC, ID52-2022), the Human Sciences Research Council Research Ethics Committee (HSRC REC, REC 2/23/09/20) and the Provincial Health Research Committee of the Department of Health of KwaZulu-Natal (KZ_202209_022) in South Africa and from the Swiss Ethics Committee Northwest and Central Switzerland (EKNZ, AO_2022–00044). This manuscript is based on protocol V.4.0, 19 January 2024. Trial findings will be disseminated through peer-reviewed publications, conference presentations and through communication offices of the consortium partners and the project’s website ( https://tbtriage.com/ ). Trial registration ClinicalTrials.gov ( NCT05526885 ), South African National Clinical Trials Register (SANCTR; DOH-27-092022-8096).
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.027 | 0.031 |
| Meta-epidemiology (narrow) | 0.006 | 0.002 |
| Meta-epidemiology (broad) | 0.007 | 0.011 |
| Bibliometrics | 0.002 | 0.002 |
| Science and technology studies | 0.001 | 0.004 |
| Scholarly communication | 0.004 | 0.003 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.003 | 0.005 |
| Insufficient payload (model declined to judge) | 0.019 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".