CoChat
Sign Up
Cited by 34Tuberculosis Research and EpidemiologyHealth Systems, Economic Evaluations, Quality of LifeCOVID-19 diagnosis using AI

Triage of Persons With Tuberculosis Symptoms Using Artificial Intelligence–Based Chest Radiograph Interpretation: A Cost-Effectiveness Analysis

Open access
Nsengiyumva, Ntwali Placide, Hussain, Hamidah, Oxlade, Olivia, Majidulla, Arman, Nazish, Ahsana, Khan, Aamir J, Menzies, Dick, Ahmad Khan, Faiz +1 more·2021·Open Forum Infectious Diseases
DOI 10.1093/ofid/ofab567PMC PMC8671604
Open this paper in CoChat — free accountOpen PDF

Abstract

Background: In settings without access to rapid expert radiographic interpretation, artificial intelligence (AI)–based chest radiograph (CXR) analysis can triage persons presenting with possible tuberculosis (TB) symptoms, to identify those who require additional microbiological testing. However, there is limited evidence of the cost-effectiveness of this technology as a triage tool. Methods: A decision analysis model was developed to evaluate the cost-effectiveness of triage strategies with AI-based CXR analysis for patients presenting with symptoms suggestive of pulmonary TB in Karachi, Pakistan. These strategies were compared to the current standard of care using microbiological testing with smear microscopy or GeneXpert, without prior triage. Positive triage CXRs were considered to improve referral success for microbiologic testing, from 91% to 100% for eligible persons. Software diagnostic accuracy was based on a prospective field study in Karachi. Other inputs were obtained from the Pakistan TB Program. The analysis was conducted from the healthcare provider perspective, and costs were expressed in 2020 US dollars. Results: Compared to upfront smear microscopy for all persons with presumptive TB, triage strategies with AI-based CXR analysis were projected to lower costs by 19%, from $23233 per 1000 persons, and avert 3%–4% disability-adjusted life-years (DALYs), from 372 DALYs. Compared to upfront GeneXpert, AI-based triage strategies lowered projected costs by 37%, from $34346 and averted 4% additional DALYs, from 369 DALYs. Reinforced follow-up for persons with positive triage CXRs but negative microbiologic tests was particularly cost-effective. Conclusions: In lower-resource settings, the addition of AI-based CXR triage before microbiologic testing for persons with possible TB symptoms can reduce costs, avert additional DALYs, and improve TB detection.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1093/ofid/ofab567

Cites

3 of 3 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Nsengiyumva et al. confirmed that AI-based CXR triage remained dominant when diagnostic specificity was maintained above 80%.

    “Nsengiyumva et al.31 confirmed that AI-based CXR triage remained dominant when diagnostic specificity was maintained above 80%”

    From Systematic review of cost effectiveness and budget impact of artificial intelligence in healthcare (2025) · cites this paper as Nsengiyumva et al.31

    The Results (Scenario and Sensitivity Analyses; page unavailable) report that AI-based CXR triage was cost-saving relative to the status quo across the probabilistic simulations. But the paper does not establish dominance conditional on specificity above 80%: the Methods (TB Epidemiology Inputs; page unavailable) list specificities of 69% and 75%, and the Discussion says specificity at the requisite sensitivity was below the WHO’s 80% target. (LLM, conf=0.88)

    Experimental — accuracy not yet measured

  • Other effectiveness metrics reported across studies included DALYs averted, diagnostic yield improvements, clinical event prevention, and years of life gained.

    “other effectiveness metrics reported across studies included DALYs averted31, diagnostic yield improvements38, clinical event prevention34, and years of life gained22”

    From Systematic review of cost effectiveness and budget impact of artificial intelligence in healthcare (2025) · cites this paper as [31]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 33

  1. [2]Sun AY, Pai M, Salje H, et al. Modeling the impact of alternative strategies for rapid molecular diagnosis of tuberculosis in Southeast Asia. Am J Epidemiol 2013; 178: 1740– 9. 24100953 PMC3858106
  2. [4]Qin ZZ, Naheyan T, Ruhwald M, et al. A new resource on artificial intelligence powered computer automated detection software products for tuberculosis programmes and implementers. Tuberculosis (Edinb) 2021; 127: 102049. 33440315
  3. [5]Harris M, Qi A, Jeagal L, et al. A systematic review of the diagnostic accuracy of artificial intelligence-based computer programs to analyze chest x-rays for pulmonary tuberculosis. PLoS One 2019; 14: e0221339. 31479448 PMC6719854
  4. [6]Khan FA, Majidulla A, Tavaziva G, et al. Chest x-ray analysis with deep learning-based software as a triage test for pulmonary tuberculosis: a prospective study of diagnostic accuracy for culture-confirmed disease. Lancet Digit Health 2020; 2: e573– 81. 33328086
  5. [7]Murphy K, Habib SS, Zaidi SMA, et al. Computer aided detection of tuberculosis on chest radiographs: an evaluation of the CAD4TB v6 system. Sci Rep 2020; 10: 5492. 32218458 PMC7099074
  6. [8]Pande T, Cohen C, Pai M, Ahmad Khan F. Computer-aided detection of pulmonary tuberculosis on digital chest radiographs: a systematic review. Int J Tuberc Lung Dis 2016; 20: 1226– 30. 27510250
  7. [9]Philipsen RH, Sánchez CI, Maduskar P, et al. Automated chest-radiography as a triage for Xpert testing in resource-constrained settings: a prospective study of diagnostic accuracy and costs. Sci Rep 2015; 5: 12215. 26212560 PMC4515744
  8. [11]Qadeer E, Fatima R, Yaqoob A, et al. Population based national tuberculosis prevalence survey among adults (>15 years) in Pakistan, 2010–2011. PLoS One 2016; 11: e0148293. 26863617 PMC4749340
  9. [12]Grzybowski S. Drugs are not enough. Failure of short-course chemotherapy in a district in India. Tuber Lung Dis 1993; 74: 145– 6. 8369506
  10. [13]Ragonnet R, Flegg JA, Brilleman SL, et al. Revisiting the natural history of pulmonary tuberculosis: a Bayesian estimation of natural recovery and mortality rates. Clin Infect Dis 2021; 73: e88– 96. 32766718
  11. [14]Tiemersma EW, van der Werf MJ, Borgdorff MW, et al. Natural history of tuberculosis: duration and fatality of untreated pulmonary tuberculosis in HIV negative patients: a systematic review. PLoS One 2011; 6: e17601. 21483732 PMC3070694
  12. [15]Ismail IM, Kibballi Madhukeshwar A, Naik PR, et al. Magnitude and reasons for gaps in tuberculosis diagnostic testing and treatment initiation: an operational research study from Dakshina Kannada, South India. J Epidemiol Glob Health 2020; 10: 326– 36. 32959619 PMC7758854
  13. [16]Grzybowski S, Enarson DA. The fate of cases of pulmonary tuberculosis under various treatment programmes. Bull IUAT 1978; 53: 70– 5. 737353
  14. [17]Minion J, Brunet L, Pai M, Fluorescent light emitting diode (LED) microscopy for the detection of Mycobacterium tuberculosis: a systematic review and meta-analysis. 2009. http://www.stoptb.org/wg/gli/assets/documents/egm%20who%20led%20report%20submitted_sept_01_2009.pdf. Accessed 17 September 2021.
  15. [20]van’t Hoog AH, Meme HK, Laserson KF, et al. Screening strategies for tuberculosis prevalence surveys: the value of chest radiography and symptoms. PLoS One 2012; 7: e38691. 22792158 PMC3391193
  16. [21]Wisnivesky JP, Kaplan J, Henschke C, et al. Evaluation of clinical parameters to predict Mycobacterium tuberculosis in inpatients. Arch Intern Med 2000; 160: 2471– 6. 10979058
  17. [24]Salomon JA, Haagsma JA, Davis A, et al. Disability weights for the Global Burden of Disease 2013 study. Lancet Glob Health 2015; 3: e712– 23. 26475018
  18. [26]Hussain H, Mori AT, Khan AJ, et al. Correction to: The cost-effectiveness of incentive-based active case finding for tuberculosis (TB) control in the private sector Karachi, Pakistan. BMC Health Serv Res 2019; 19: 794. 31690293 PMC6829995
  19. [32]Ochalek J, Lomas J, Claxton K. Estimating health opportunity costs in low-income and middle-income countries: a novel approach and evidence from cross-country data. BMJ Glob Health 2018; 3: e000964. PMC6231096 30483412
  20. [33]World Health Organization. Macroeconomics and Health: Investing in Health for Economic Development: Report of the Commission on Macroeconomics and Health. Geneva, Switzerland: WHO; 2001: 202.
  21. [35]Houben RMGJ, Lalli M, Kranzer K, et al. What if they don’t have tuberculosis? The consequences and trade-offs involved in false-positive diagnoses of tuberculosis. Clin Infect Dis 2019; 68: 150– 6. 29982375 PMC6293007
  22. [36]Kendall EA, Kamoga C, Kitonsa PJ, et al. Empiric treatment of pulmonary TB in the Xpert era: correspondence of sputum culture, Xpert MTB/RIF, and clinical diagnoses. PLoS One 2019; 14: e0220251. 31339935 PMC6655770
  23. [37]McCarthy K, Fielding K, Churchyard GJ, Grant AD. Empiric tuberculosis treatment in South African primary health care facilities—for whom, where, when and why: implications for the development of tuberculosis diagnostic tests. PLoS One 2018; 13: e0191608. 29364960 PMC5783417
  24. [38]Breuninger M, van Ginneken B, Philipsen RH, et al. Diagnostic accuracy of computer-aided detection of pulmonary tuberculosis in chest radiographs: a validation study from sub-Saharan Africa. PLoS One 2014; 9: e106381. 25192172 PMC4156349
  25. [39]Melendez J, Sánchez CI, Philipsen RH, et al. An automated tuberculosis screening strategy combining X-ray-based computer-aided detection and clinical information. Sci Rep 2016; 6: 25265. 27126741 PMC4850474
  26. [40]Muyoyeta M, Maduskar P, Moyo M, et al. The sensitivity and specificity of using a computer aided diagnosis program for automatically scoring chest X-rays of presumptive TB patients compared with Xpert MTB/RIF in Lusaka, Zambia. PLoS One 2014; 9: e93757. 24705629 PMC3976315
  27. [41]Rahman MT, Codlin AJ, Rahman MM, et al. An evaluation of automated chest radiography reading software for tuberculosis screening among public-and private-sector patients. Eur Respir J 2017; 49: 1602159. 28529202 PMC5460641
  28. [42]Zaidi SMA, Habib SS, Van Ginneken B, et al. Evaluation of the diagnostic accuracy of computer-aided detection of tuberculosis on chest radiography among private sector patients in Pakistan. Sci Rep 2018; 8: 12339. 30120345 PMC6098114
  29. [43]Qin ZZ, Sander MS, Rai B, et al. Using artificial intelligence to read chest radiographs for tuberculosis detection: a multi-site evaluation of the diagnostic accuracy of three deep learning systems. Sci Rep 2019; 9: 15000. 31628424 PMC6802077
  30. [44]Koesoemadinata RC, Kranzer K, Livia R, et al. Computer-assisted chest radiography reading for tuberculosis screening in people living with diabetes mellitus. Int J Tuberc Lung Dis 2018; 22: 1088– 94. 30092877
  31. [45]Melendez J, Hogeweg L, Sánchez CI, et al. Accuracy of an automated system for tuberculosis detection on chest radiographs in high-risk screening. Int J Tuberc Lung Dis 2018; 22: 567– 71. 29663963 PMC5905390
  32. [46]Melendez J, Philipsen RHHM, Chanda-Kapata P, et al. Automatic versus human reading of chest X-rays in the Zambia National Tuberculosis Prevalence Survey. Int J Tuberc Lung Dis 2017; 21: 880– 6. 28786796
  33. [47]Muyoyeta M, Kasese NC, Milimo D, et al. Digital CXR with computer aided diagnosis versus symptom screen to define presumptive tuberculosis among household contacts and impact on tuberculosis diagnosis. BMC Infect Dis 2017; 17: 301. 28438139 PMC5402643

Cited by

Sorted by citations

Authors
Nsengiyumva, N. P., Hussain, H., Oxlade, O., Majidulla, A., Nazish, A., Khan, A. J., Menzies, D., Ahmad Khan, F., & Schwartzman, K.
Affiliations
  • McGill University Health Centre (CA)
  • Interactive Research and Development (PK)
  • Indus Hospital (PK)
  • McGill International Tuberculosis Centre, Research Institute of the McGill University Health Centre, Montreal, Canada
Corresponding author
Schwartzman, Kevin
ORCID
Venue
Open Forum Infectious Diseases
Publisher
Oxford University Press (OUP)
Published
2021-12-01
Volume
8
Issue
12
Pages
ofab567-ofab567
ISSN
2328-8957 (ISSN-L)
DOI
10.1093/ofid/ofab567
PMC
PMC8671604
OpenAlex
W4200434450
Cited by
34
References
3
Citations by year
2026: 12 · 2025: 11 · 2024: 7 · 2023: 4
FWCI
0.87
Type
journal article
Language
en
Open access
Licence
Locations
  • Open Forum Infectious Diseases (publishedVersion, cc-by-nc-nd)
  • PubMed Central (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
Funders
Canadian Institutes of Health Research (PJT-148743)
Topics
  • Health Sciences › Medicine › Infectious Diseases › Tuberculosis Research and Epidemiology
  • Social Sciences › Economics, Econometrics and Finance › Economics and Econometrics › Health Systems, Economic Evaluations, Quality of Life
  • Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › COVID-19 diagnosis using AI
Keywords
tuberculosis, artificial intelligence, chest X-ray interpretation, triage, cost-effectiveness, GeneXpert, smear microscopy, microbiological analysis, disability-adjusted life years, tuberculosis detection, Karachi

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review