Evaluation of an ultra-portable X-ray system with automated interpretation for tuberculosis active case finding in carceral settings: a diagnostic test accuracy study

Evaluation of an ultra-portable X-ray system with automated interpretation for tuberculosis active case finding in carceral settings: a diagnostic test accuracy study | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Evaluation of an ultra-portable X-ray system with automated interpretation for tuberculosis active case finding in carceral settings: a diagnostic test accuracy study Argita D. Salindri, José V. B. Bampi, Caroline Busatto, Alessandra M. da Silva, and 10 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-5578367/v1 This work is licensed under a CC BY 4.0 License Status: Published Journal Publication published 03 Nov, 2025 Read the published version in BMC Infectious Diseases → Version 1 posted 11 You are reading this latest preprint version Abstract Background The World Health Organization recommends systematic active case finding for tuberculosis (TB) among high-risk population including incarcerated individuals; however, many prisons lack screening capacity. In this study, we aimed to evaluate the diagnostic performance of an ultra-portable digital chest radiography system paired with LunitTB, an automated interpretation algorithm, to detect TB disease. Methods We performed a diagnostic test accuracy study using data collected for a prospective active case finding study for TB in a Brazilian prison from February 2023 through May 2024. Eligible individuals included adults (≥18 years) without a TB history in the past two years. A Fujifilm Digital Radiography (FDR) Xair paired with LunitTB algorithm (version v3.1.5.1) system was used to screen consented individuals for TB disease irrespective of their TB symptoms. Area under curve (AUC) and 95% confidence intervals (CI) were estimated to determine the accuracy of FDR Xair and LunitTB interpretation when compared to a rigorous microbiologic reference standard. Results We screened a total of 3409 individuals for TB disease as part of our active TB case finding study, and 3399 (99.7%) met our eligibility criteria for the diagnostic test accuracy study. TB prevalence was 4.1% (139/3399, 95%CI 3.5–4.8%). The AUC for FDR Xair and LunitTB interpretation was 0.89 (0.86–0.93). The accuracy of FDR Xair and LunitTB interpretation among those with any TB symptoms was significantly higher (AUC = 0.93, 95%CI 0.90–0.97) compared to those without TB symptoms (AUC = 0.87, 95%CI 0.81–0.92) (DeLong p = 0.033). Conclusions The FDR Xair and LunitTB interpretation enabled us to screen persons deprived of liberty rapidly, with a high diagnostic accuracy especially among those reported any TB symptoms. active case finding persons deprived of liberty ultra-portable X-ray LunitTB Figures Figure 1 Figure 2 Figure 3 Figure 4 INTRODUCTION Globally, persons deprived of liberty (PDL) have exceedingly high risk of tuberculosis (TB); especially in South American countries where the incidence rate was estimated to be nearly 27-fold than the general population. [ 1 ] In 2021, the World Health Organization (WHO) recommended systematic screening of PDL for TB disease. [ 2 ] However, implementation of this recommendation has been limited in low- and middle-income countries, as many carceral institutions lack resources and equipment for systematic TB screening. There is a critical need to identify efficient strategies for TB active case finding that can be scaled in carceral settings in high TB burden countries. In recent years, there have major advances in the use of artificial intelligence to interpret chest X-ray (CXR) images for TB screening, with promising results; however, the majority of published studies were conducted among symptomatic individuals presenting to clinical settings. [ 3 ] Recent technological development in digital X-ray imaging and display have provided an opportunity to bringing care outside of healthcare facilities through portable X-ray devices. [ 4 ] The use of portable X-ray system paired with an automated X-ray interpretations was recently shown to be effective for community-based TB screening in an evaluation in Nigeria. [ 5 ] Furthermore, a recent systematic review aiming to evaluate the performance of different CXR paired with AI software reported a pooled sensitivity of 94% (89–96%) and a pooled specificity of 95% (91–97%) in model-development studies (i.e., non-trial studies). [ 6 ] However, the accuracy of ultra-portable radiography with automated interpretation compared with a rigorous microbiologic reference standard remains unclear. Thus, we aimed to evaluate the diagnostic performance of an ultra-portable X-ray device and automated interpretation system as a screening test for active TB case finding efforts among the prison population. METHODS Study population, design, and setting We conducted a diagnostic test accuracy study using data collected for a prospective TB active case finding study in a large male prison in the state of Mato Grosso do Sul, Brazil, from February, 2023 through May, 2024. For the active TB case finding study, all adult PDL (³18 years) were approached; and after obtaining informed consent, study staff administered structured demographic and clinical questionnaires. All study participants were then asked to provide a spot sputum sample, which was divided for a) pooled GeneXpert Ultra (Xpert) with a pool size of eight as previously described [7, 8] and b) Ogawa culture testing. By the end of enrollment day, all sputum samples were transported to the local public health reference laboratory where all microbiological works was performed. After sputum collection, a posterior-anterior chest X-ray was obtained using Fujifilm Digital Radiography (FDR) Xair XD2000 PX for all participants irrespective of TB symptoms. FDR Xair is a lightweight digital radiography system. X-ray images were scored using LunitTB algorithms v3.1.5.1 developed by Lunit (Seoul, South Korea), which provides a numerical TB risk score between 0 and 100. For this diagnostic test accuracy study, eligible participants included consented individuals with no history of TB treatment in the past two years. Study measures, definitions, and statistical analysis Demographic characteristics and clinical data were collected using a structured study questionnaire (Supplemental Material 1) developed for the present study’s purposes and recorded using REDCap [9] online data capture tools hosted at the Federal University of Mato Grosso do Sul. Collected information included age, highest education attainment, incarceration history, previous TB history, TB symptoms at the time of TB screening, and behavioral risk factors (e.g., smoking and drug use). We categorized smoking status into “current smoker” and “never/former smoker.” We define any drug use if study participants reported any drug use in the past 12-months period. Individuals with a positive Xpert result at screening had an Xpert confirmatory test done; thus, we defined our study outcome, TB disease, if individuals had a) a positive culture result, b) two positive Xpert test results, or c) one positive and one trace Xpert results. [10] Statistical Analyses We used chi-square and Fisher’s exact tests to assess bivariate associations between participants’ characteristics and TB status. We used Wilcoxon rank sum test to compare the median of LunitTB scores among individuals with and without TB disease. We then estimated the area under the receiver operating characteristic (ROC) curve (AUC) and the 95% confidence interval (CI) to quantify the accuracy of FDR Xair and LunitTB screening system as a quantitative diagnostic for TB disease. We performed sensitivity analyses to evaluate the performance of FDR Xair and LunitTB screening system when different definitions were used to define TB (i.e., a positive result on both culture and Xpert, and a positive result on either culture or Xpert). We also calculated sensitivity at the LunitTB threshold that achieved 70% specificity (LunitTB score=42.8), and specificity at 90% sensitivity (LunitTB score=32.7), corresponding to the WHO screening benchmarks. [2] We used DeLong test [11] to compare areas under correlated ROC curves among key sub-populations. We compared LunitTB scores and evaluated sensitivity according to the semi-quantitative Xpert result, as a measure of bacillary burden. Ethics approval and consent to participate The study was approved by the Research Ethics Committee of the Federal University of Mato Grosso do Sul (#5.730.361), the National Research Ethics Committee of Brazil (CONEP) (#5.899.470), and the Institutional Review Boards (IRB) at Stanford University (IRB#67287). All study participants provided written informed consent prior to study participation and study procedures were performed in accordance with relevant guidelines and regulations. RESULTS Characteristics of study participants and TB prevalence We screened 3409 individuals for TB disease as part of our active TB case finding study in 85 working days, for a median of 40 individuals screened per day and a maximum of 96. Among these, 3408 (99.9%) had LunitTB score and TB status information available, nine of whom had a TB episode within two years prior to study enrollment and were excluded, leaving 3399 (99.7%) individuals included in the final analyses (Fig. 1) . Characteristics of individuals screened are provided in Table 1 . Overall, TB prevalence was 4.1% (139/3399, 95%CI 3.5–4.8) (Table 1) . Compared to individuals without TB, those with TB were more likely to have a history of TB treatment > 2 years prior to study enrollment, report at least one TB symptom, and currently smoke tobacco products (p < 0.05). The median LunitTB score was significantly higher among individuals with TB (median = 96.7, interquartile range [IQR] 90.0–98.4) compared to those without TB (median = 29.0, IQR 15.0–49.1, p-value < 0.001). Samples of scored chest X-ray images from individuals with and without TB disease are provided in Fig. 2 . Diagnostic accuracy of the FDR Xair and LunitTB interpretation Overall, the area under curve (AUC) for TB prediction using FDR Xair and LunitTB interpretation was 0.89 (95%CI 0.86–0.93) (Fig. 3) . Using a positive Xpert and a positive culture to define TB, the AUC was improved to 0.93 (95%CI 0.90–0.96) (p = 0.076). The AUC was lower when using a positive Xpert or a positive culture to define TB (AUC = 0.84, 95%CI 0.80–0.88) (p = 0.048). The diagnostic accuracy among those with any TB symptoms (AUC = 0.93, 95%CI 0.90–0.97) was significantly higher compared to those without TB symptoms (AUC = 0.87, 95%CI 0.81–0.92) (p = 0.033) (Fig. 4A) . Similarly, diagnostic accuracy was significantly higher among current smokers (AUC = 0.92, 95%CI 0.89–0.95) compared to never/former smokers (AUC = 0.80, 95%CI 0.69–0.90) (p = 0.028). The diagnostic accuracy was similar among individuals with or without history of TB episode > 2 years prior to study enrollment (p = 0.552). Among individuals with any TB symptoms, at 70% specificity, the screening system met the WHO target product profile (TPP) [ 2 ] thresholds for a screening test with a sensitivity of 92.5% (95%CI 84.9–98.1) (Fig. 4A) . Similarly, among individuals with any TB symptoms, at 90% sensitivity, the FDR Xair and LunitTB interpretation met the WHO thresholds for a screening test with a specificity of 89.1% (95%CI 52.6–94.5) (data not shown). Considerations to use FDR Xair and LunitTB interpretation in an active TB case finding effort Nearly a third (32.3%, 1099/3399) of study participants had LunitTB score > 42.8 (data not shown). Performing individual Xpert tests among those with LunitTB score > 42.8, we would have used 68% fewer Xpert cartridges (i.e., compared to performing individual Xpert tests among all study participants; 1099 vs. 3399 cartridges) while maintaining sensitivity to identify a high proportion (123/139, 88.5%) of individuals with TB disease (Fig. 4A) . Among the 16 individuals with TB and LunitTB score ≤42.8 that we would have missed without individual Xpert testing, 2 had positive culture and negative Xpert results, while 14 others had either low (n = 5), very low (n = 8) or trace Xpert (n = 1) test results (Fig. 4 B ). LunitTB scores were strongly correlated with Xpert semi-quantitative load (Spearman’s ρ = 0.387, p < 0.001), and all 47 (100%) study participants with medium (n = 23) or high (n = 24) bacterial load were identified by using the 42.8 LunitTB score threshold (Fig. 4B) . Among our study participants, there were 578 (17.0%, 578/3399) individuals who reported at least one TB symptoms, 220 (38.1%) of whom had LunitTB score > 42.8. Performing individual Xpert and/or culture tests (i.e., microbiologic reference standard) among those with LunitTB score > 42.8 and ≥1 TB symptoms, we would have identified approximately one-third (35.3%, 49/139) of all TB cases per our study definition (data not shown). Symptoms screening alone with Xpert confirmation would have detected 38.1% (95%CI 30.3–46.4%, 53/139) of TB cases. DISCUSSION Overall, the FDR Xair and LunitTB interpretation enabled efficient TB active case finding to be performed among PDL and met the WHO TPP benchmarks, achieving particularly high accuracy among individuals with any TB symptoms. Furthermore, the FDR-Xair-LunitTB system was sensitive in identifying PDL with medium/high bacterial load, who may be more likely to transmit Mtb . Incorporating FDR Xair and LunitTB interpretation in active case finding programs in prisons could reduce the number of Xpert cartridges used while identifying individuals at highest risk for morbidity and transmission. While the WHO recommends use of chest radiography for TB active case finding, many institutional correctional or detention facilities in high TB burden countries do not have X-ray equipment or have equipment that are not functioning [ 12 ]. Repairs are often delayed or not performed due to need for technical personnel to travel to prisons and obtained security clearances. We previously used a mobile X-ray machine installed on a container truck for TB screening in prisons, finding high accuracy of its use combined with automated interpretation for TB screening [ 13 ]. However, this approach required transportation of PDL from the cell blocks to the exterior courtyard of the prison, which required substantial security personnel time and slowed the pace of screening. The ultra-portable, battery-powered, FDR Xair–LunitTB system enabled screening to be done within the cell blocks, achieving more rapid screening, and can be transported between prisons to increase access to screening while containing costs. It requires low to no constructions costs, produced high quality images [ 14 ], can be repaired off-site, and has acceptable radiation risks that can be enhanced with portable lead curtains [ 15 ]. The accuracy of the Xair–LunitTB screening system was higher among individuals with TB symptoms. Individuals with symptomatic TB were more likely to have higher Xpert semiquantitative loads than those with subclinical TB, and we believe the greater accuracy of the radiographic screening reflects greater sensitivity in more advanced TB disease. Screening sensitivity was moderately high (86% at 70% specificity) among those without symptoms, but this fell short of WHO TPP benchmarks. It is critical to note that only one-third of our study participants reported ≥1 TB symptoms. The use of the ultra-portable X-ray with LunitTB interpretation among symptomatic individuals would have only detected one-third of TB cases in the prison. As individuals with subclinical TB (i.e., asymptomatic) may play an important role in transmission, these findings suggest a need to further evaluate the consideration of lowering the threshold especially among this sub-group or to identify alternative tools to improve screening sensitivity. Our finding also suggested that approximately one out of nine (~ 11.5%) TB cases in our study would not have been identified using CXR and LunitTB system alone. The use of CXR and LunitTB in combination with other screening strategies (e.g., TB symptoms assessment and Xpert confirmation) may improve screening yields as documented in our previous study. [ 16 ] Future studies should also consider evaluating the added value of periodically rescreening individuals with abnormal CXR findings at baseline screening as these individuals may be at greater risks to develop TB disease. [ 17 ] Our study has several limitations. First, our study was conducted in a single maximum-security prison in Brazil, and the results, may not be generalizable to other settings with different demographic and epidemiologic characteristics. Specifically, the LunitTB score thresholds identified in this population should not be extrapolated to other study populations (e.g., household contacts) as TB risk factors (e.g., smoking, poor ventilations) may be more common among PDL. [ 18 ] Second, we used spot sputum samples for Xpert and/or culture testing, which may lead to misclassification of TB status. Third, we did not perform a proper costing analysis (i.e., cost-effectiveness analysis) to measure how much screening costs were saved by using the ultra-portable X-ray and LunitTB interpretation compared to other screening strategies. However, our previous work suggested that screening strategies which utilized the combination of symptoms screening, chest X-ray with an automated interpretation, and Xpert confirmation had a slightly higher yield (76% vs. 74% when using sputum Xpert for all participants) with an average cost per case diagnosed of US $ 395. [ 16 ] Fourth, we did not perform a validity study to compare the AI reading with a radiologist reading to interpret CXR images. Fifth, our study was conducted in settings where HIV prevalence was low (~ 1%) [ 19 ]. Thus, we did not do any stratifications according to HIV status due to the small sample size (n = 14) as it will be challenging to make any inferences on HIV-related findings. CONCLUSION In conclusion, the FDR Xair and LunitTB interpretation enabled us to screen PDL rapidly, with a high diagnostic accuracy especially among individuals reporting any TB symptoms. Further studies to assess the performance of FDR Xair and LunitTB interpretation in other settings are still warranted. Abbreviations AUC Area under the curve CI Confidence interval FDR Fujifilm digital radiography IQR Interquartile range PDL Persons deprived of Liberty ROC Receiver operating characteristics TB Tuberculosis TPP Target Product Profile WHO World Health Organization Declarations Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request. Competing interest We have no conflict of interest to declare. Funding This work was supported in part by grants from the National Institutes of Health (NIH) including the National Institute of Allergy and Infectious Diseases (NIAID) [R01AI130058 to JRA and JC]. Authors contributions JRA and JC conceived the study design. JVBB, CB, AMdS, and IBG collected data. AdSS, TOG, and EATC, assisted with the sample collections/processing and data collection process. ADS performed the analyses. ADS, JVBB, JRA, and JC interpreted the results. ADS wrote the first draft of the manuscript. JRA and JC assisted with further drafting and revisions of the manuscript. All authors reviewed and approved the final version of the manuscript. Acknowledgements The authors thank the State Agency of Administration Prisons (AGEPEN) for their full support during the study period, the study participants for their kind cooperation during the data collection process, and the Central Laboratory (LACEN) of the state of Mato Grosso do Sul for the support in the accomplishment of the laboratory tests. References Cords O, et al. Incidence and prevalence of tuberculosis in incarcerated populations: a systematic review and meta-analysis. Lancet Public Health. 2021;6(5):e300–8. WHO consolidated guidelines on tuberculosis: Module 2: screening – systematic screening for tuberculosis disease . 2021. Harris M, 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(9):e0221339. Henderson D et al. Portable X-rays–A new era? IPEM-Translation, 2022. 3–4: p. 100005. John S, et al. Comparing tuberculosis symptom screening to chest X-ray with artificial intelligence in an active case finding campaign in Northeast Nigeria. BMC Global Public Health. 2023;1(1):17. Zhan Y et al. Diagnostic Accuracy of the Artificial Intelligence Methods in Medical Imaging for Pulmonary Tuberculosis: A Systematic Review and Meta-Analysis. J Clin Med, 2022. 12(1). Dos Santos PCP, et al. Pooling Sputum Samples for Efficient Mass Tuberculosis Screening in Prisons. Clin Infect Dis. 2022;74(12):2115–21. Batestin D, et al. Sputum pooling for rapid and cost-effective active case-finding for TB in prisons. Int J Tuberc Lung Dis. 2025;29(2):88–9. Harris PA, et al. The REDCap consortium: Building an international community of software platform partners. J Biomed Inf. 2019;95:103208. Crowder R et al. Head-to-head comparison of diagnostic accuracy of TB screening tests: Chest-X-ray, Xpert TB host response, and C-reactive protein. medRxiv, 2024. DeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44(3):837–45. Charalambous S et al. Scaling up evidence-based approaches to tuberculosis screening in prisons. Lancet Public Health, 2023. Soares TR, et al. Evaluation of chest X-ray with automated interpretation algorithms for mass tuberculosis screening in prisons: a cross-sectional study. Lancet Reg Health Am. 2023;17:100388. Kamal R, et al. A comparison of the quality of images of chest X-ray between handheld portable digital X-ray & routinely used digital X-ray machine. Indian J Med Res. 2023;157(23):204–10. Vo LNQ et al. Early Evaluation of an Ultra-Portable X-ray System for Tuberculosis Active Case Finding. Trop Med Infect Dis, 2021. 6(3). Santos ADS, et al. Yield, Efficiency, and Costs of Mass Screening Algorithms for Tuberculosis in Brazilian Prisons. Clin Infect Dis. 2021;72(5):771–7. Tan Q, et al. Chest Radiograph Screening for Detecting Subclinical Tuberculosis in Asymptomatic Household Contacts, Peru. Emerg Infect Dis. 2024;30(6):1115–24. Qin ZZ, 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(1):15000. Sgarbi RV, et al. A Cross-Sectional Survey of HIV Testing and Prevalence in Twelve Brazilian Correctional Facilities. PLoS ONE. 2015;10(10):e0139487. Tables Table 1. Characteristics and risk factors for prevalent tuberculosis among persons deprived of liberty in Mato Grosso do Sul, Brazil, February 2023 – May 2024 (N=3,399) Characteristics Total N = 3399 Tuberculosis Status * p-values † No N (%) = 3260 (95.9) Yes N (%) = 139 (4.1) Median age, years (IQR) 31 (26 – 37) 31 (26 – 37) 30 (26 – 36) 0.4985 Race White 634 (18.7) 617 (18.9) 17 (12.2) 0.032 § Black 292 (8.6) 279 (8.6) 13 (9.4) Mixed 2466 (72.6) 2358 (72.3) 108 (77.7) Asian 4 (0.1) 4 (0.1) 0 (0.0) Indigenous 3 (0.1) 2 (0.1) 1 (0.7) Education attainment Did not complete high school 2908 (85.6) 2781 (85.3) 127 (91.4) 0.047 Completed high school 491 (14.4) 479 (14.7) 12 (8.6) Previously incarcerated 2837 (83.5) 2714 (83.3) 123 (88.5) 0.104 Smoking status Never/former smoker Current smoker 1320 (38.8) 2079 (61.2) 1288 (39.5) 1972 (60.5) 32 (23.0) 107 (77.0) <0.001 Any drug use Marijuana Cocaine Crack Heroin Glue and/or other solvents Pasta-based Hashish Injectables 2286 (67.3) 2072/2286 (90.6) 1113/2286 (48.7) 55/2286 (2.4) 10/2286 (0.4) 18/2286 (0.8) 62/2286 (2.7) 27/2286 (1.2) 2/2286 (0.1) 2183 (67.0) 1978/2183 (90.6) 1061/2183 (48.6) 53/2183 (2.4) 10/2183 (0.5) 17/2183 (0.8) 59/2183 (2.7) 27/2183 (1.2) 2/2183 (0.1) 103 (74.1) 94/103 (91.3) 52/103 (50.5) 2/103 (1.9) 0/103 (0.0) 1/103 (1.0) 3/103 (2.9) 0/103 (0.0) 0/103 (0.0) 0.079 0.824 0.709 1.000 § 1.000 § 0.565 § 0.757 § 0.631 § 1.000 § Previous TB a 574 (16.9) 534 (16.4) 40 (28.8) <0.001 Any TB symptoms Cough Productive cough Blood-stained sputum Fever Loss of appetite Weight loss Night sweats Chest pain Difficulty in breathing 578 (17.0) 474/578 (82.0) 375/578 (64.9) 35/578 (6.1) 170/578 (29.4) 86/578 (14.9) 145/578 (25.1) 91/578 (15.7) 215/578 (37.2) 182/578 (31.5) 525 (16.1) 423/525 (80.6) 333/525 (63.4) 28/525 (5.3) 146/525 (27.8) 71/525 (13.5) 126/525 (24.0) 77/525 (14.7) 183/525 (34.9) 158/525 (30.1) 53 (38.1) 51/53 (96.2) 42/53 (79.2) 7/53 (13.2) 24/53 (45.3) 15/53 (28.3) 19/53 (35.8) 14/53 (26.4) 32/53 (60.4) 24/53 (45.3) <0.001 0.005 0.022 0.012 § 0.008 0.004 0.058 0.025 <0.001 0.023 Contact with TB-sick person (N=3398) Yes, 1 – 3 times per week Yes, 4 – 6 times per week Yes, everyday 2044 (60.2) 146/2044 (7.1) 42/2044 (2.1) 1856/2044 (90.8) 1962 (60.2) 142/1961 (7.2) 41/1961 (2.1) 1779/1961 (90.7) 82 (59.0) 4/83 (4.9) 1/83 (1.2) 77/83 (93.9) 0.775 0.608 LunitTB score, median (IQR) 29.9 (15.0 – 53.4) 29.0 (15.0 – 49.1) 96.7 (89.9 – 98.4) <0.001 ‡ * Tuberculosis case was defined by a positive culture or two positive GeneXpert or one positive and one trace GeneXpert results † p-values from Chi-square tests, unless indicated otherwise ‡ p-values from Wilcoxon rank-sum tests § p-values from Fisher’s exact tests a Previous TB episode dated >2 years prior to study enrollment Abbreviations IQR – interquartile range; TB – tuberculosis Bold indicates that the finding was statistically significant at a=0.05 Additional Declarations No competing interests reported. Supplementary Files CompleteStudyQuestionnairesENGTranslation.pdf Cite Share Download PDF Status: Published Journal Publication published 03 Nov, 2025 Read the published version in BMC Infectious Diseases → Version 1 posted Editor assigned by journal 21 Apr, 2025 Editorial decision: Revision requested 18 Apr, 2025 Reviews received at journal 16 Apr, 2025 Reviews received at journal 16 Apr, 2025 Reviewers agreed at journal 16 Apr, 2025 Reviews received at journal 16 Apr, 2025 Reviewers agreed at journal 07 Apr, 2025 Reviewers agreed at journal 03 Apr, 2025 Reviewers invited by journal 03 Apr, 2025 Submission checks completed at journal 03 Apr, 2025 First submitted to journal 02 Apr, 2025 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-5578367","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":438105431,"identity":"9bc17f33-1a6f-4667-9e62-8aaecaf97e82","order_by":0,"name":"Argita D. Salindri","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA60lEQVRIie3PsQrCMBCA4ZNAp2jXSlFfIVIoin2Yk0K7CPoIgoNjVwUfQhCcUzK4CK4BHZSCUwdBEEEHo45K1M0hP2Q4yMdxACbTX4YAWwaF5DEE6nEA5yNBRcZ9ooboawJQmPJviT2MdwfsbSqz1So99jDoljhJ11RDnGXuOcj23lyGxB1h1BxzK2zpCMgOKCLac0nALZ4FY5z6ro7UZJyd72SWCHKheCf2SUuYRP+xZQqh5T4JtbSkvsz9hiLeSIZ+i2LEysLymhMNqS7iTB6uopIkabamGLDSYrCTue7818hv300mk8n0phso01HIcATIDAAAAABJRU5ErkJggg==","orcid":"","institution":"Stanford University School of Medicine","correspondingAuthor":true,"prefix":"","firstName":"Argita","middleName":"D.","lastName":"Salindri","suffix":""},{"id":438105432,"identity":"a6336036-a92f-446a-931b-6e4c8edc3877","order_by":1,"name":"José V. B. Bampi","email":"","orcid":"","institution":"Federal University of Mato Grosso do Sul, Mato Grosso do Sul","correspondingAuthor":false,"prefix":"","firstName":"José","middleName":"V. B.","lastName":"Bampi","suffix":""},{"id":438105433,"identity":"b8fdb776-c5d2-497c-b072-56f0afb51f30","order_by":2,"name":"Caroline Busatto","email":"","orcid":"","institution":"Federal University of Mato Grosso do Sul, Mato Grosso do Sul","correspondingAuthor":false,"prefix":"","firstName":"Caroline","middleName":"","lastName":"Busatto","suffix":""},{"id":438105434,"identity":"e5656833-f7a1-4811-be79-15c423dcd4d0","order_by":3,"name":"Alessandra M. da Silva","email":"","orcid":"","institution":"Federal University of Mato Grosso do Sul, Mato Grosso do Sul","correspondingAuthor":false,"prefix":"","firstName":"Alessandra","middleName":"M. da","lastName":"Silva","suffix":""},{"id":438105435,"identity":"0ea58f0e-9bce-445e-b58b-98475d2d924c","order_by":4,"name":"Andrea da Silva Santos","email":"","orcid":"","institution":"Federal University of Grande Dourados","correspondingAuthor":false,"prefix":"","firstName":"Andrea","middleName":"da Silva","lastName":"Santos","suffix":""},{"id":438105436,"identity":"ad471cea-168d-48a5-a28d-c4bd2a14c823","order_by":5,"name":"Isabella B. Gonçalves","email":"","orcid":"","institution":"Federal University of Mato Grosso do Sul, Mato Grosso do Sul","correspondingAuthor":false,"prefix":"","firstName":"Isabella","middleName":"B.","lastName":"Gonçalves","suffix":""},{"id":438105437,"identity":"017d2597-0c80-4790-8e69-8fce08510217","order_by":6,"name":"Thais O. Gonçalves","email":"","orcid":"","institution":"Central Laboratory of Mato Gross do Sul","correspondingAuthor":false,"prefix":"","firstName":"Thais","middleName":"O.","lastName":"Gonçalves","suffix":""},{"id":438105438,"identity":"41a3a2da-43a7-468d-98ce-ba24b70ba44f","order_by":7,"name":"Eunice A. T. Cunha","email":"","orcid":"","institution":"Central Laboratory of Mato Gross do Sul","correspondingAuthor":false,"prefix":"","firstName":"Eunice","middleName":"A. T.","lastName":"Cunha","suffix":""},{"id":438105439,"identity":"5b9cbca1-0893-4a63-b290-20072d2b9486","order_by":8,"name":"Daniel Tsuha","email":"","orcid":"","institution":"Federal University of Mato Grosso do Sul, Mato Grosso do Sul","correspondingAuthor":false,"prefix":"","firstName":"Daniel","middleName":"","lastName":"Tsuha","suffix":""},{"id":438105440,"identity":"7f87778b-4e0c-4ee0-97ea-f2060f90d4d2","order_by":9,"name":"Everton Lemos","email":"","orcid":"","institution":"Federal University of Mato Grosso do Sul, Mato Grosso do Sul","correspondingAuthor":false,"prefix":"","firstName":"Everton","middleName":"","lastName":"Lemos","suffix":""},{"id":438105441,"identity":"3887dcc1-0f1a-4679-ba2b-0e8a5b6a7642","order_by":10,"name":"Roberto D. Oliveira","email":"","orcid":"","institution":"Federal University of Grande Dourados","correspondingAuthor":false,"prefix":"","firstName":"Roberto","middleName":"D.","lastName":"Oliveira","suffix":""},{"id":438105442,"identity":"9d5ae866-a94b-4041-b4f5-a41f30b702ba","order_by":11,"name":"Mariana Croda","email":"","orcid":"","institution":"Federal University of Mato Grosso do Sul, Mato Grosso do Sul","correspondingAuthor":false,"prefix":"","firstName":"Mariana","middleName":"","lastName":"Croda","suffix":""},{"id":438105443,"identity":"b8344087-e8f8-48a8-8d42-d4ec5834a534","order_by":12,"name":"Jason R. Andrews","email":"","orcid":"","institution":"Stanford University School of Medicine","correspondingAuthor":false,"prefix":"","firstName":"Jason","middleName":"R.","lastName":"Andrews","suffix":""},{"id":438105444,"identity":"c3900771-00ba-4424-a058-dc380bbd3337","order_by":13,"name":"Julio Croda","email":"","orcid":"","institution":"Federal University of Mato Grosso do Sul, Mato Grosso do Sul","correspondingAuthor":false,"prefix":"","firstName":"Julio","middleName":"","lastName":"Croda","suffix":""}],"badges":[],"createdAt":"2024-12-04 09:23:09","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-5578367/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-5578367/v1","draftVersion":[],"editorialEvents":[{"content":"https://doi.org/10.1186/s12879-025-11835-0","type":"published","date":"2025-11-03T15:57:32+00:00"}],"editorialNote":"","failedWorkflow":false,"files":[{"id":80062773,"identity":"b2055c74-a343-4c5b-b7c6-0cb10f72137a","added_by":"auto","created_at":"2025-04-07 12:40:24","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":487130,"visible":true,"origin":"","legend":"\u003cp\u003eStudy participants selection and tuberculosis case definition\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5578367/v1/4d557b685e1db8d47b861b77.jpeg"},{"id":80061816,"identity":"547f8638-0916-4375-a5ee-9da29a2df149","added_by":"auto","created_at":"2025-04-07 12:32:24","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":365588,"visible":true,"origin":"","legend":"\u003cp\u003eChest X-ray images produced with FDR Xair and scored with LunitTB for two individuals with tuberculosis (A and B) and two individuals without tuberculosis (C and D)\u003c/p\u003e","description":"","filename":"2.png","url":"https://assets-eu.researchsquare.com/files/rs-5578367/v1/c97cec252544732e9a4c7148.png"},{"id":80061823,"identity":"fbccc9e1-4d5e-4b4c-9c24-781ad623e27c","added_by":"auto","created_at":"2025-04-07 12:32:24","extension":"jpeg","order_by":3,"title":"Figure 3","display":"","copyAsset":false,"role":"figure","size":483295,"visible":true,"origin":"","legend":"\u003cp\u003eReceiver operating characteristic (ROC) curves and area under the curve (AUC) for LunitTB as a screening test for active TB case finding efforts among persons deprived of liberty in Mato Grosso do Sul, Brazil, February 2023 – May 2024 (N=3,399)\u003c/p\u003e","description":"","filename":"floatimage3.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5578367/v1/dd6094b87071811be5e1350b.jpeg"},{"id":80062775,"identity":"5a650aa7-74ae-448c-aa53-a0bffec5ea50","added_by":"auto","created_at":"2025-04-07 12:40:24","extension":"jpeg","order_by":4,"title":"Figure 4","display":"","copyAsset":false,"role":"figure","size":930080,"visible":true,"origin":"","legend":"\u003cp\u003eSee image above for figure legend\u003c/p\u003e","description":"","filename":"floatimage4.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-5578367/v1/a033e4ba970ea647be423d49.jpeg"},{"id":95564046,"identity":"782c9d52-6a73-49bc-af54-bfdd8d10c938","added_by":"auto","created_at":"2025-11-10 16:06:57","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":3117226,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5578367/v1/8ede3725-6e6d-424f-a7a2-7dc9db42562e.pdf"},{"id":80063929,"identity":"432b07ef-31a9-40b4-9e35-dd9d93f32898","added_by":"auto","created_at":"2025-04-07 12:48:24","extension":"pdf","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":501784,"visible":true,"origin":"","legend":"","description":"","filename":"CompleteStudyQuestionnairesENGTranslation.pdf","url":"https://assets-eu.researchsquare.com/files/rs-5578367/v1/e260c78c6a40b9bf09f9085f.pdf"}],"financialInterests":"No competing interests reported.","formattedTitle":"\u003cp\u003eEvaluation of an ultra-portable X-ray system with automated interpretation for tuberculosis active case finding in carceral settings: a diagnostic test accuracy study\u003c/p\u003e","fulltext":[{"header":"INTRODUCTION","content":"\u003cp\u003eGlobally, persons deprived of liberty (PDL) have exceedingly high risk of tuberculosis (TB); especially in South American countries where the incidence rate was estimated to be nearly 27-fold than the general population. [\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e] In 2021, the World Health Organization (WHO) recommended systematic screening of PDL for TB disease. [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] However, implementation of this recommendation has been limited in low- and middle-income countries, as many carceral institutions lack resources and equipment for systematic TB screening. There is a critical need to identify efficient strategies for TB active case finding that can be scaled in carceral settings in high TB burden countries.\u003c/p\u003e \u003cp\u003eIn recent years, there have major advances in the use of artificial intelligence to interpret chest X-ray (CXR) images for TB screening, with promising results; however, the majority of published studies were conducted among symptomatic individuals presenting to clinical settings. [\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e] Recent technological development in digital X-ray imaging and display have provided an opportunity to bringing care outside of healthcare facilities through portable X-ray devices. [\u003cspan citationid=\"CR4\" class=\"CitationRef\"\u003e4\u003c/span\u003e] The use of portable X-ray system paired with an automated X-ray interpretations was recently shown to be effective for community-based TB screening in an evaluation in Nigeria. [\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e] Furthermore, a recent systematic review aiming to evaluate the performance of different CXR paired with AI software reported a pooled sensitivity of 94% (89\u0026ndash;96%) and a pooled specificity of 95% (91\u0026ndash;97%) in model-development studies (i.e., non-trial studies). [\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e] However, the accuracy of ultra-portable radiography with automated interpretation compared with a rigorous microbiologic reference standard remains unclear. Thus, we aimed to evaluate the diagnostic performance of an ultra-portable X-ray device and automated interpretation system as a screening test for active TB case finding efforts among the prison population.\u003c/p\u003e"},{"header":"METHODS","content":"\u003cp\u003e\u003cem\u003eStudy population, design, and setting\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eWe conducted a diagnostic test accuracy study using data collected for a prospective TB active case finding study in a large male prison in the state of Mato Grosso do Sul, Brazil, from February, 2023 through May, 2024. For the active TB case finding study, all adult PDL (\u0026sup3;18 years) were approached; and after obtaining informed consent, study staff administered structured demographic and clinical questionnaires. All study participants were then asked to provide a spot sputum sample, which was divided for a) pooled GeneXpert Ultra (Xpert) with a pool size of eight as previously described [7, 8] and b) Ogawa culture testing. By the end of enrollment day, all sputum samples were transported to the local public health reference laboratory where all microbiological works was performed. After sputum collection, a posterior-anterior chest X-ray was obtained using Fujifilm Digital Radiography (FDR) Xair XD2000 PX for all participants irrespective of TB symptoms. FDR Xair is a lightweight digital radiography system. X-ray images were scored using LunitTB algorithms v3.1.5.1 developed by Lunit (Seoul, South Korea), which provides a numerical TB risk score between 0 and 100. For this diagnostic test accuracy study, eligible participants included consented individuals with no history of TB treatment in the past two years.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eStudy measures, definitions, and statistical analysis\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eDemographic characteristics and clinical data were collected using a structured study questionnaire \u003cstrong\u003e(Supplemental Material 1)\u003c/strong\u003e developed for the present study\u0026rsquo;s purposes and recorded using REDCap [9] online data capture tools hosted at the Federal University of Mato Grosso do Sul. Collected information included age, highest education attainment, incarceration history, previous TB history, TB symptoms at the time of TB screening, and behavioral risk factors (e.g., smoking and drug use). We categorized smoking status into \u0026ldquo;current smoker\u0026rdquo; and \u0026ldquo;never/former smoker.\u0026rdquo; We define any drug use if study participants reported any drug use in the past 12-months period.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eIndividuals with a positive Xpert result at screening had an Xpert confirmatory test done; thus, we defined our study outcome, TB disease, if individuals had a) a positive culture result, b) two positive Xpert test results, or c) one positive and one trace Xpert results. [10]\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eStatistical Analyses\u003c/em\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eWe used chi-square and Fisher\u0026rsquo;s exact tests to assess bivariate associations between participants\u0026rsquo; characteristics and TB status. We used Wilcoxon rank sum test to compare the median of LunitTB scores among individuals with and without TB disease. We then estimated the area under the receiver operating characteristic (ROC) curve (AUC) and the 95% confidence interval (CI) to quantify the accuracy of FDR Xair and LunitTB screening system as a quantitative diagnostic for TB disease. We performed sensitivity analyses to evaluate the performance of FDR Xair and LunitTB screening system when different definitions were used to define TB (i.e., a positive result on both culture and Xpert, and a positive result on either culture or Xpert). We also calculated sensitivity at the LunitTB threshold that achieved 70% specificity (LunitTB score=42.8), and specificity at 90% sensitivity (LunitTB score=32.7), corresponding to the WHO screening benchmarks. [2] We used DeLong test [11] to compare areas under correlated ROC curves among key sub-populations. We compared LunitTB scores and evaluated sensitivity according to the semi-quantitative Xpert result, as a measure of bacillary burden.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eEthics approval and consent to participate\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eThe study was approved by the Research Ethics Committee of the Federal University of Mato Grosso do Sul (#5.730.361), the National Research Ethics Committee of Brazil (CONEP) (#5.899.470), and the Institutional Review Boards (IRB) at Stanford University (IRB#67287). All study participants provided written informed consent prior to study participation and study procedures were performed in accordance with relevant guidelines and regulations.\u003c/p\u003e"},{"header":"RESULTS","content":" \u003cp\u003e \u003cem\u003eCharacteristics of study participants and TB prevalence\u003c/em\u003e \u003c/p\u003e \u003cp\u003eWe screened 3409 individuals for TB disease as part of our active TB case finding study in 85 working days, for a median of 40 individuals screened per day and a maximum of 96. Among these, 3408 (99.9%) had LunitTB score and TB status information available, nine of whom had a TB episode within two years prior to study enrollment and were excluded, leaving 3399 (99.7%) individuals included in the final analyses \u003cb\u003e(Fig.\u0026nbsp;1)\u003c/b\u003e. Characteristics of individuals screened are provided in \u003cb\u003eTable\u0026nbsp;1\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eOverall, TB prevalence was 4.1% (139/3399, 95%CI 3.5–4.8) \u003cb\u003e(Table\u0026nbsp;1)\u003c/b\u003e. Compared to individuals without TB, those with TB were more likely to have a history of TB treatment \u0026gt; 2 years prior to study enrollment, report at least one TB symptom, and currently smoke tobacco products (p \u0026lt; 0.05). The median LunitTB score was significantly higher among individuals with TB (median = 96.7, interquartile range [IQR] 90.0–98.4) compared to those without TB (median = 29.0, IQR 15.0–49.1, p-value \u0026lt; 0.001). Samples of scored chest X-ray images from individuals with and without TB disease are provided in \u003cb\u003eFig.\u0026nbsp;2\u003c/b\u003e.\u003c/p\u003e \u003cp\u003e \u003cem\u003eDiagnostic accuracy of the FDR Xair and LunitTB interpretation\u003c/em\u003e \u003c/p\u003e \u003cp\u003eOverall, the area under curve (AUC) for TB prediction using FDR Xair and LunitTB interpretation was 0.89 (95%CI 0.86–0.93) \u003cb\u003e(Fig.\u0026nbsp;3)\u003c/b\u003e. Using a positive Xpert and a positive culture to define TB, the AUC was improved to 0.93 (95%CI 0.90–0.96) (p = 0.076). The AUC was lower when using a positive Xpert or a positive culture to define TB (AUC = 0.84, 95%CI 0.80–0.88) (p = 0.048).\u003c/p\u003e \u003cp\u003eThe diagnostic accuracy among those with any TB symptoms (AUC = 0.93, 95%CI 0.90–0.97) was significantly higher compared to those without TB symptoms (AUC = 0.87, 95%CI 0.81–0.92) (p = 0.033) \u003cb\u003e(Fig.\u0026nbsp;4A)\u003c/b\u003e. Similarly, diagnostic accuracy was significantly higher among current smokers (AUC = 0.92, 95%CI 0.89–0.95) compared to never/former smokers (AUC = 0.80, 95%CI 0.69–0.90) (p = 0.028). The diagnostic accuracy was similar among individuals with or without history of TB episode \u0026gt; 2 years prior to study enrollment (p = 0.552).\u003c/p\u003e \u003cp\u003eAmong individuals with any TB symptoms, at 70% specificity, the screening system met the WHO target product profile (TPP) [\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e] thresholds for a screening test with a sensitivity of 92.5% (95%CI 84.9–98.1) \u003cb\u003e(Fig.\u0026nbsp;4A)\u003c/b\u003e. Similarly, among individuals with any TB symptoms, at 90% sensitivity, the FDR Xair and LunitTB interpretation met the WHO thresholds for a screening test with a specificity of 89.1% (95%CI 52.6–94.5) (data not shown).\u003c/p\u003e \u003cp\u003e \u003cem\u003eConsiderations to use FDR Xair and LunitTB interpretation in an active TB case finding effort\u003c/em\u003e \u003c/p\u003e \u003cp\u003eNearly a third (32.3%, 1099/3399) of study participants had LunitTB score \u0026gt; 42.8 (data not shown). Performing individual Xpert tests among those with LunitTB score \u0026gt; 42.8, we would have used 68% fewer Xpert cartridges (i.e., compared to performing individual Xpert tests among all study participants; 1099 vs. 3399 cartridges) while maintaining sensitivity to identify a high proportion (123/139, 88.5%) of individuals with TB disease \u003cb\u003e(Fig.\u0026nbsp;4A)\u003c/b\u003e. Among the 16 individuals with TB and LunitTB score ≤42.8 that we would have missed without individual Xpert testing, 2 had positive culture and negative Xpert results, while 14 others had either low (n = 5), very low (n = 8) or trace Xpert (n = 1) test results (Fig.\u0026nbsp;4\u003cb\u003eB\u003c/b\u003e). LunitTB scores were strongly correlated with Xpert semi-quantitative load (Spearman’s ρ = 0.387, p \u0026lt; 0.001), and all 47 (100%) study participants with medium (n = 23) or high (n = 24) bacterial load were identified by using the 42.8 LunitTB score threshold \u003cb\u003e(Fig.\u0026nbsp;4B)\u003c/b\u003e.\u003c/p\u003e \u003cp\u003eAmong our study participants, there were 578 (17.0%, 578/3399) individuals who reported at least one TB symptoms, 220 (38.1%) of whom had LunitTB score \u0026gt; 42.8. Performing individual Xpert and/or culture tests (i.e., microbiologic reference standard) among those with LunitTB score \u0026gt; 42.8 and ≥1 TB symptoms, we would have identified approximately one-third (35.3%, 49/139) of all TB cases per our study definition (data not shown). Symptoms screening alone with Xpert confirmation would have detected 38.1% (95%CI 30.3–46.4%, 53/139) of TB cases.\u003c/p\u003e"},{"header":"DISCUSSION","content":"\u003cp\u003eOverall, the FDR Xair and LunitTB interpretation enabled efficient TB active case finding to be performed among PDL and met the WHO TPP benchmarks, achieving particularly high accuracy among individuals with any TB symptoms. Furthermore, the FDR-Xair-LunitTB system was sensitive in identifying PDL with medium/high bacterial load, who may be more likely to transmit \u003cem\u003eMtb\u003c/em\u003e. Incorporating FDR Xair and LunitTB interpretation in active case finding programs in prisons could reduce the number of Xpert cartridges used while identifying individuals at highest risk for morbidity and transmission.\u003c/p\u003e\u003cp\u003eWhile the WHO recommends use of chest radiography for TB active case finding, many institutional correctional or detention facilities in high TB burden countries do not have X-ray equipment or have equipment that are not functioning [\u003cspan citationid=\"CR12\" class=\"CitationRef\"\u003e12\u003c/span\u003e]. Repairs are often delayed or not performed due to need for technical personnel to travel to prisons and obtained security clearances. We previously used a mobile X-ray machine installed on a container truck for TB screening in prisons, finding high accuracy of its use combined with automated interpretation for TB screening [\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e]. However, this approach required transportation of PDL from the cell blocks to the exterior courtyard of the prison, which required substantial security personnel time and slowed the pace of screening. The ultra-portable, battery-powered, FDR Xair–LunitTB system enabled screening to be done within the cell blocks, achieving more rapid screening, and can be transported between prisons to increase access to screening while containing costs. It requires low to no constructions costs, produced high quality images [\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e], can be repaired off-site, and has acceptable radiation risks that can be enhanced with portable lead curtains [\u003cspan citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e].\u003c/p\u003e\u003cp\u003eThe accuracy of the Xair–LunitTB screening system was higher among individuals with TB symptoms. Individuals with symptomatic TB were more likely to have higher Xpert semiquantitative loads than those with subclinical TB, and we believe the greater accuracy of the radiographic screening reflects greater sensitivity in more advanced TB disease. Screening sensitivity was moderately high (86% at 70% specificity) among those without symptoms, but this fell short of WHO TPP benchmarks. It is critical to note that only one-third of our study participants reported ≥1 TB symptoms. The use of the ultra-portable X-ray with LunitTB interpretation among symptomatic individuals would have only detected one-third of TB cases in the prison. As individuals with subclinical TB (i.e., asymptomatic) may play an important role in transmission, these findings suggest a need to further evaluate the consideration of lowering the threshold especially among this sub-group or to identify alternative tools to improve screening sensitivity.\u003c/p\u003e\u003cp\u003eOur finding also suggested that approximately one out of nine (~ 11.5%) TB cases in our study would not have been identified using CXR and LunitTB system alone. The use of CXR and LunitTB in combination with other screening strategies (e.g., TB symptoms assessment and Xpert confirmation) may improve screening yields as documented in our previous study. [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] Future studies should also consider evaluating the added value of periodically rescreening individuals with abnormal CXR findings at baseline screening as these individuals may be at greater risks to develop TB disease. [\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e]\u003c/p\u003e\u003cp\u003eOur study has several limitations. First, our study was conducted in a single maximum-security prison in Brazil, and the results, may not be generalizable to other settings with different demographic and epidemiologic characteristics. Specifically, the LunitTB score thresholds identified in this population should not be extrapolated to other study populations (e.g., household contacts) as TB risk factors (e.g., smoking, poor ventilations) may be more common among PDL. [\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e] Second, we used spot sputum samples for Xpert and/or culture testing, which may lead to misclassification of TB status. Third, we did not perform a proper costing analysis (i.e., cost-effectiveness analysis) to measure how much screening costs were saved by using the ultra-portable X-ray and LunitTB interpretation compared to other screening strategies. However, our previous work suggested that screening strategies which utilized the combination of symptoms screening, chest X-ray with an automated interpretation, and Xpert confirmation had a slightly higher yield (76% vs. 74% when using sputum Xpert for all participants) with an average cost per case diagnosed of US\u003cspan\u003e$\u003c/span\u003e395. [\u003cspan citationid=\"CR16\" class=\"CitationRef\"\u003e16\u003c/span\u003e] Fourth, we did not perform a validity study to compare the AI reading with a radiologist reading to interpret CXR images. Fifth, our study was conducted in settings where HIV prevalence was low (~ 1%) [\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e]. Thus, we did not do any stratifications according to HIV status due to the small sample size (n = 14) as it will be challenging to make any inferences on HIV-related findings.\u003c/p\u003e"},{"header":"CONCLUSION","content":"\u003cp\u003eIn conclusion, the FDR Xair and LunitTB interpretation enabled us to screen PDL rapidly, with a high diagnostic accuracy especially among individuals reporting any TB symptoms. Further studies to assess the performance of FDR Xair and LunitTB interpretation in other settings are still warranted.\u003c/p\u003e"},{"header":"Abbreviations","content":"\u003cp\u003eAUC\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Area under the curve\u003c/p\u003e\n\u003cp\u003eCI \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Confidence interval\u003c/p\u003e\n\u003cp\u003eFDR \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Fujifilm digital radiography\u003c/p\u003e\n\u003cp\u003eIQR\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Interquartile range\u003c/p\u003e\n\u003cp\u003ePDL\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Persons deprived of Liberty\u003c/p\u003e\n\u003cp\u003eROC\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Receiver operating characteristics\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eTB\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;Tuberculosis\u003c/p\u003e\n\u003cp\u003eTPP\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;Target Product Profile\u003c/p\u003e\n\u003cp\u003eWHO \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;World Health Organization\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interest\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe have no conflict of interest to declare.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThis work was supported in part by grants from the National Institutes of Health (NIH) including the National Institute of Allergy and Infectious Diseases (NIAID) [R01AI130058 to JRA and JC]. \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJRA and JC conceived the study design. JVBB, CB, AMdS, and IBG collected data. AdSS, TOG, and EATC, assisted with the sample collections/processing and data collection process. ADS performed the analyses. ADS, JVBB, JRA, and JC interpreted the results. ADS wrote the first draft of the manuscript. JRA and JC assisted with further drafting and revisions of the manuscript. All authors reviewed and approved the final version of the manuscript.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe authors thank the State Agency of Administration Prisons (AGEPEN) for their full support during the study period, the study participants for their kind cooperation during the data collection process, and the Central Laboratory (LACEN) of the state of Mato Grosso do Sul for the support in the accomplishment of the laboratory tests.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\u003cli\u003e\u003cspan\u003eCords O, et al. Incidence and prevalence of tuberculosis in incarcerated populations: a systematic review and meta-analysis. Lancet Public Health. 2021;6(5):e300\u0026ndash;8.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003e\u003cem\u003eWHO consolidated guidelines on tuberculosis: Module 2: screening \u0026ndash; systematic screening for tuberculosis disease\u003c/em\u003e. 2021.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarris M, 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(9):e0221339.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHenderson D et al. Portable X-rays\u0026ndash;A new era? IPEM-Translation, 2022. 3\u0026ndash;4: p. 100005.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eJohn S, et al. Comparing tuberculosis symptom screening to chest X-ray with artificial intelligence in an active case finding campaign in Northeast Nigeria. BMC Global Public Health. 2023;1(1):17.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eZhan Y et al. Diagnostic Accuracy of the Artificial Intelligence Methods in Medical Imaging for Pulmonary Tuberculosis: A Systematic Review and Meta-Analysis. J Clin Med, 2022. 12(1).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDos Santos PCP, et al. Pooling Sputum Samples for Efficient Mass Tuberculosis Screening in Prisons. Clin Infect Dis. 2022;74(12):2115\u0026ndash;21.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eBatestin D, et al. Sputum pooling for rapid and cost-effective active case-finding for TB in prisons. Int J Tuberc Lung Dis. 2025;29(2):88\u0026ndash;9.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eHarris PA, et al. The REDCap consortium: Building an international community of software platform partners. J Biomed Inf. 2019;95:103208.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCrowder R et al. \u003cem\u003eHead-to-head comparison of diagnostic accuracy of TB screening tests: Chest-X-ray, Xpert TB host response, and C-reactive protein.\u003c/em\u003e medRxiv, 2024.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eDeLong ER, DeLong DM, Clarke-Pearson DL. Comparing the areas under two or more correlated receiver operating characteristic curves: a nonparametric approach. Biometrics. 1988;44(3):837\u0026ndash;45.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eCharalambous S et al. Scaling up evidence-based approaches to tuberculosis screening in prisons. Lancet Public Health, 2023.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSoares TR, et al. Evaluation of chest X-ray with automated interpretation algorithms for mass tuberculosis screening in prisons: a cross-sectional study. Lancet Reg Health Am. 2023;17:100388.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eKamal R, et al. A comparison of the quality of images of chest X-ray between handheld portable digital X-ray \u0026amp; routinely used digital X-ray machine. Indian J Med Res. 2023;157(23):204\u0026ndash;10.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eVo LNQ et al. Early Evaluation of an Ultra-Portable X-ray System for Tuberculosis Active Case Finding. Trop Med Infect Dis, 2021. 6(3).\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSantos ADS, et al. Yield, Efficiency, and Costs of Mass Screening Algorithms for Tuberculosis in Brazilian Prisons. Clin Infect Dis. 2021;72(5):771\u0026ndash;7.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eTan Q, et al. Chest Radiograph Screening for Detecting Subclinical Tuberculosis in Asymptomatic Household Contacts, Peru. Emerg Infect Dis. 2024;30(6):1115\u0026ndash;24.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eQin ZZ, 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(1):15000.\u003c/span\u003e\u003c/li\u003e \u003cli\u003e\u003cspan\u003eSgarbi RV, et al. A Cross-Sectional Survey of HIV Testing and Prevalence in Twelve Brazilian Correctional Facilities. PLoS ONE. 2015;10(10):e0139487.\u003c/span\u003e\u003c/li\u003e \u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003e\u003cstrong\u003eTable 1.\u0026nbsp;\u003c/strong\u003eCharacteristics and risk factors for prevalent tuberculosis among persons deprived of liberty in Mato Grosso do Sul, Brazil, February 2023 \u0026ndash; May 2024 (N=3,399)\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"790\"\u003e\n \u003cthead\u003e\n \u003ctr\u003e\n \u003ctd rowspan=\"2\" style=\"width: 271px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristics\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 138px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTotal\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eN = 3399\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd colspan=\"2\" valign=\"top\" style=\"width: 292px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eTuberculosis Status\u003csup\u003e*\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd rowspan=\"2\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u003cstrong\u003ep-values\u003csup\u003e\u0026dagger;\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNo\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eN (%) = 3260 (95.9)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eYes\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eN (%) = 139 (4.1)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/thead\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eMedian age, years (IQR)\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e31 (26 \u0026ndash; 37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e31 (26 \u0026ndash; 37)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e30 (26 \u0026ndash; 36)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.4985\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eRace\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;White\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e634 (18.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e617 (18.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e17 (12.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.032\u003csup\u003e\u0026sect;\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Black\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e292 (8.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e279 (8.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e13 (9.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Mixed\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e2466 (72.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e2358 (72.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e108 (77.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Asian\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e4 (0.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e4 (0.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e0 (0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Indigenous\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e3 (0.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e2 (0.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e1 (0.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eEducation attainment\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Did not complete high school\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e2908 (85.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e2781 (85.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e127 (91.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e0.047\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Completed high school\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e491 (14.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e479 (14.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e12 (8.6)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003ePreviously incarcerated\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e2837 (83.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e2714 (83.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e123 (88.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.104\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eSmoking status\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Never/former smoker\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Current smoker\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e1320 (38.8)\u003c/p\u003e\n \u003cp\u003e2079 (61.2)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e1288 (39.5)\u003c/p\u003e\n \u003cp\u003e1972 (60.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e32 (23.0)\u003c/p\u003e\n \u003cp\u003e107 (77.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026nbsp;\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eAny drug use\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Marijuana\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Cocaine\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Crack\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Heroin\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Glue and/or other solvents\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Pasta-based\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Hashish\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Injectables\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e2286 (67.3)\u003c/p\u003e\n \u003cp\u003e2072/2286 (90.6)\u003c/p\u003e\n \u003cp\u003e1113/2286 (48.7)\u003c/p\u003e\n \u003cp\u003e55/2286 (2.4)\u003c/p\u003e\n \u003cp\u003e10/2286 (0.4)\u003c/p\u003e\n \u003cp\u003e18/2286 (0.8)\u003c/p\u003e\n \u003cp\u003e62/2286 (2.7)\u003c/p\u003e\n \u003cp\u003e27/2286 (1.2)\u003c/p\u003e\n \u003cp\u003e2/2286 (0.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e2183 (67.0)\u003c/p\u003e\n \u003cp\u003e1978/2183 (90.6)\u003c/p\u003e\n \u003cp\u003e1061/2183 (48.6)\u003c/p\u003e\n \u003cp\u003e53/2183 (2.4)\u003c/p\u003e\n \u003cp\u003e10/2183 (0.5)\u003c/p\u003e\n \u003cp\u003e17/2183 (0.8)\u003c/p\u003e\n \u003cp\u003e59/2183 (2.7)\u003c/p\u003e\n \u003cp\u003e27/2183 (1.2)\u003c/p\u003e\n \u003cp\u003e2/2183 (0.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e103 (74.1)\u003c/p\u003e\n \u003cp\u003e94/103 (91.3)\u003c/p\u003e\n \u003cp\u003e52/103 (50.5)\u003c/p\u003e\n \u003cp\u003e2/103 (1.9)\u003c/p\u003e\n \u003cp\u003e0/103 (0.0)\u003c/p\u003e\n \u003cp\u003e1/103 (1.0)\u003c/p\u003e\n \u003cp\u003e3/103 (2.9)\u003c/p\u003e\n \u003cp\u003e0/103 (0.0)\u003c/p\u003e\n \u003cp\u003e0/103 (0.0)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.079\u003c/p\u003e\n \u003cp\u003e0.824\u003c/p\u003e\n \u003cp\u003e0.709\u003c/p\u003e\n \u003cp\u003e1.000\u003csup\u003e\u0026sect;\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003e1.000\u003csup\u003e\u0026sect;\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003e0.565\u003csup\u003e\u0026sect;\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003e0.757\u003csup\u003e\u0026sect;\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003e0.631\u003csup\u003e\u0026sect;\u003c/sup\u003e\u003c/p\u003e\n \u003cp\u003e1.000\u003csup\u003e\u0026sect;\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003ePrevious TB\u003csup\u003ea\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e574 (16.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e534 (16.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e40 (28.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eAny TB symptoms\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Cough\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Productive cough\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Blood-stained sputum\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Fever\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Loss of appetite\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Weight loss\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Night sweats\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Chest pain\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Difficulty in breathing\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e578 (17.0)\u003c/p\u003e\n \u003cp\u003e474/578 (82.0)\u003c/p\u003e\n \u003cp\u003e375/578 (64.9)\u003c/p\u003e\n \u003cp\u003e35/578 (6.1)\u003c/p\u003e\n \u003cp\u003e170/578 (29.4)\u003c/p\u003e\n \u003cp\u003e86/578 (14.9)\u003c/p\u003e\n \u003cp\u003e145/578 (25.1)\u003c/p\u003e\n \u003cp\u003e91/578 (15.7)\u003c/p\u003e\n \u003cp\u003e215/578 (37.2)\u003c/p\u003e\n \u003cp\u003e182/578 (31.5)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e525 (16.1)\u003c/p\u003e\n \u003cp\u003e423/525 (80.6)\u003c/p\u003e\n \u003cp\u003e333/525 (63.4)\u003c/p\u003e\n \u003cp\u003e28/525 (5.3)\u003c/p\u003e\n \u003cp\u003e146/525 (27.8)\u003c/p\u003e\n \u003cp\u003e71/525 (13.5)\u003c/p\u003e\n \u003cp\u003e126/525 (24.0)\u003c/p\u003e\n \u003cp\u003e77/525 (14.7)\u003c/p\u003e\n \u003cp\u003e183/525 (34.9)\u003c/p\u003e\n \u003cp\u003e158/525 (30.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e53 (38.1)\u003c/p\u003e\n \u003cp\u003e51/53 (96.2)\u003c/p\u003e\n \u003cp\u003e42/53 (79.2)\u003c/p\u003e\n \u003cp\u003e7/53 (13.2)\u003c/p\u003e\n \u003cp\u003e24/53 (45.3)\u003c/p\u003e\n \u003cp\u003e15/53 (28.3)\u003c/p\u003e\n \u003cp\u003e19/53 (35.8)\u003c/p\u003e\n \u003cp\u003e14/53 (26.4)\u003c/p\u003e\n \u003cp\u003e32/53 (60.4)\u003c/p\u003e\n \u003cp\u003e24/53 (45.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0.005\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0.022\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0.012\u003csup\u003e\u0026sect;\u003c/sup\u003e\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0.008\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0.004\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e0.058\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0.025\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.001\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003e0.023\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eContact with TB-sick person (N=3398)\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes, 1 \u0026ndash; 3 times per week\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes, 4 \u0026ndash; 6 times per week\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;Yes, everyday\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e2044 (60.2)\u003c/p\u003e\n \u003cp\u003e146/2044 (7.1)\u003c/p\u003e\n \u003cp\u003e42/2044 (2.1)\u003c/p\u003e\n \u003cp\u003e1856/2044 (90.8)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e1962 (60.2)\u003c/p\u003e\n \u003cp\u003e142/1961 (7.2)\u003c/p\u003e\n \u003cp\u003e41/1961 (2.1)\u003c/p\u003e\n \u003cp\u003e1779/1961 (90.7)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e82 (59.0)\u003c/p\u003e\n \u003cp\u003e4/83 (4.9)\u003c/p\u003e\n \u003cp\u003e1/83 (1.2)\u003c/p\u003e\n \u003cp\u003e77/83 (93.9)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e0.775\u003c/p\u003e\n \u003cp\u003e0.608\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 271px;\"\u003e\n \u003cp\u003eLunitTB score, median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 138px;\"\u003e\n \u003cp\u003e29.9 (15.0 \u0026ndash; 53.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 151px;\"\u003e\n \u003cp\u003e29.0 (15.0 \u0026ndash; 49.1)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 140px;\"\u003e\n \u003cp\u003e96.7 (89.9 \u0026ndash; 98.4)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 89px;\"\u003e\n \u003cp\u003e\u003cstrong\u003e\u0026lt;0.001\u003c/strong\u003e\u003csup\u003e\u0026Dagger;\u003c/sup\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd colspan=\"5\" valign=\"top\" style=\"width: 790px;\"\u003e\n \u003cp\u003e\u003csup\u003e*\u003c/sup\u003eTuberculosis case was defined by a positive culture or two positive GeneXpert or one positive and one trace GeneXpert results\u003c/p\u003e\n \u003cp\u003e\u003csup\u003e\u0026dagger;\u003c/sup\u003ep-values from Chi-square tests, unless indicated otherwise\u003c/p\u003e\n \u003cp\u003e\u003csup\u003e\u0026Dagger;\u003c/sup\u003ep-values from Wilcoxon rank-sum tests\u003c/p\u003e\n \u003cp\u003e\u003csup\u003e\u0026sect;\u003c/sup\u003ep-values from Fisher\u0026rsquo;s exact tests\u003c/p\u003e\n \u003cp\u003e\u003csup\u003ea\u003c/sup\u003ePrevious TB episode dated \u0026gt;2 years prior to study enrollment\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eAbbreviations\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003eIQR \u0026ndash; interquartile range; TB \u0026ndash; tuberculosis\u003c/p\u003e\n \u003cp\u003e\u003cstrong\u003eBold\u0026nbsp;\u003c/strong\u003eindicates that the finding was statistically significant at a=0.05\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":true,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"bmc-infectious-diseases","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"infd","sideBox":"Learn more about [BMC Infectious Diseases](http://bmcinfectdis.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/infd","title":"BMC Infectious Diseases","twitterHandle":"#bmcinfectdis","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"active case finding, persons deprived of liberty, ultra-portable X-ray, LunitTB","lastPublishedDoi":"10.21203/rs.3.rs-5578367/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-5578367/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003ch2\u003eBackground\u003c/h2\u003e \u003cp\u003eThe World Health Organization recommends systematic active case finding for tuberculosis (TB) among high-risk population including incarcerated individuals; however, many prisons lack screening capacity. In this study, we aimed to evaluate the diagnostic performance of an ultra-portable digital chest radiography system paired with LunitTB, an automated interpretation algorithm, to detect TB disease.\u003c/p\u003e\u003ch2\u003eMethods\u003c/h2\u003e \u003cp\u003eWe performed a diagnostic test accuracy study using data collected for a prospective active case finding study for TB in a Brazilian prison from February 2023 through May 2024. Eligible individuals included adults (\u0026ge;18 years) without a TB history in the past two years. A Fujifilm Digital Radiography (FDR) Xair paired with LunitTB algorithm (version v3.1.5.1) system was used to screen consented individuals for TB disease irrespective of their TB symptoms. Area under curve (AUC) and 95% confidence intervals (CI) were estimated to determine the accuracy of FDR Xair and LunitTB interpretation when compared to a rigorous microbiologic reference standard.\u003c/p\u003e\u003ch2\u003eResults\u003c/h2\u003e \u003cp\u003eWe screened a total of 3409 individuals for TB disease as part of our active TB case finding study, and 3399 (99.7%) met our eligibility criteria for the diagnostic test accuracy study. TB prevalence was 4.1% (139/3399, 95%CI 3.5\u0026ndash;4.8%). The AUC for FDR Xair and LunitTB interpretation was 0.89 (0.86\u0026ndash;0.93). The accuracy of FDR Xair and LunitTB interpretation among those with any TB symptoms was significantly higher (AUC\u0026thinsp;=\u0026thinsp;0.93, 95%CI 0.90\u0026ndash;0.97) compared to those without TB symptoms (AUC\u0026thinsp;=\u0026thinsp;0.87, 95%CI 0.81\u0026ndash;0.92) (DeLong p\u0026thinsp;=\u0026thinsp;0.033).\u003c/p\u003e\u003ch2\u003eConclusions\u003c/h2\u003e \u003cp\u003eThe FDR Xair and LunitTB interpretation enabled us to screen persons deprived of liberty rapidly, with a high diagnostic accuracy especially among those reported any TB symptoms.\u003c/p\u003e","manuscriptTitle":"Evaluation of an ultra-portable X-ray system with automated interpretation for tuberculosis active case finding in carceral settings: a diagnostic test accuracy study","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-04-07 12:32:19","doi":"10.21203/rs.3.rs-5578367/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorAssigned","content":"","date":"2025-04-21T20:21:57+00:00","index":"","fulltext":""},{"type":"decision","content":"Revision requested","date":"2025-04-18T13:05:46+00:00","index":"","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-16T16:57:44+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-16T15:11:22+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"236959285313578568791924540795650299343","date":"2025-04-16T13:29:52+00:00","index":"hide","fulltext":""},{"type":"editorInvitedReview","content":"","date":"2025-04-16T08:35:53+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"94114429444159783683698145626383518396","date":"2025-04-07T07:02:56+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"185740407433124153675916818585997674744","date":"2025-04-03T18:12:29+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2025-04-03T17:16:39+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2025-04-03T07:45:19+00:00","index":"","fulltext":""},{"type":"submitted","content":"BMC Infectious Diseases","date":"2025-04-02T11:19:28+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"bmc-infectious-diseases","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"infd","sideBox":"Learn more about [BMC Infectious Diseases](http://bmcinfectdis.biomedcentral.com/)","snPcode":"","submissionUrl":"https://www.editorialmanager.com/infd","title":"BMC Infectious Diseases","twitterHandle":"#bmcinfectdis","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"em","reportingPortfolio":"BMC Series","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"6acdd45d-e217-4a3f-b2b9-d6282a9fe6b0","owner":[],"postedDate":"April 7th, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"published-in-journal","subjectAreas":[],"tags":[],"updatedAt":"2025-11-10T16:01:18+00:00","versionOfRecord":{"articleIdentity":"rs-5578367","link":"https://doi.org/10.1186/s12879-025-11835-0","journal":{"identity":"bmc-infectious-diseases","isVorOnly":false,"title":"BMC Infectious Diseases"},"publishedOn":"2025-11-03 15:57:32","publishedOnDateReadable":"November 3rd, 2025"},"versionCreatedAt":"2025-04-07 12:32:19","video":"","vorDoi":"10.1186/s12879-025-11835-0","vorDoiUrl":"https://doi.org/10.1186/s12879-025-11835-0","workflowStages":[]},"version":"v1","identity":"rs-5578367","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-5578367","identity":"rs-5578367","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}