Impact of Regional Collaborative Network on the Rescue of Patients with ST-Elevation Myocardial Infarction First Visit to non-PCI Hospitals.

Objective: To assess the effect of a regional collaborative network on the treatment of ST-elevation myocardial infarction (STEMI) patients first admitted to non- percutaneous coronary intervention (PCI) hospitals. Methods: Using data from Kunshan Hospital of Traditional Chinese Medicine's chest pain center database, patients were grouped based on the establishment of the regional collaborative rescue network. Key timepoints and in-hospital complications were analyzed. Results: A total of 152 ST-elevation myocardial infarction patients were included in the study. Compared to control group, symptom-to-balloon time (S-B), time of first medical contact to balloon and inter-hospital referral time in observation group were significantly shorter [(314.03 ± 209.26) min vs (451.27 ± 290.44) min, P = .001], [(115.32 ± 54.73) min vs (191.67 ± 130.30) min, P = .001], [(55.09 ± 37.23) min vs (112.67 ± 95.90) min, P = .001], but time of symptom to first medical contact were not statistically significant[(210.27±217.07) min vs (239.61 ± 200.92) min, P = .136].The incidence of heart failure and total complications during hospitalization decreased [7 (8.14%) vs 13 (19.70%), P = .037] and [14 (16.28%) vs 24 (36.36%), P = .004]. However no statistically significant difference were observed in rate of death during hospitalization [2 (2.33%) vs 3 (4.55%), P = .450], ventricular fibrillation [2 (2.33%) vs 3 (4.55%), P = .450], left ventricular thrombosis [2 (2.33%) vs 4 (6.06%), P = .244] and recurrent myocardial infarction[1 (1.16%) vs 1 (1.52%), P = .851]. Conclusions: The regional cooperative rescue network notably reduces ischemic and referral times for STEMI patients, lowering the incidence of heart failure during their hospital stay.

Folic Acid Protects Against Kidney Damage in Mice with Diabetic Nephropathy by Inhibiting M1 Macrophage Polarization via Nuclear Factor-k-gene Binding Pathway.

Kidney damage is one of the most common complications of diabetes, and inflammation caused by macrophage infiltration plays an important role. Folic acid (FA), a water-soluble vitamin, was previously found to affect inflammation by regulating macrophage polarization. In our study, we aimed to investigate the effect of FA on renal injury in mice with diabetic nephropathy (DN). We found that FA treatment ameliorated diabetic metabolic parameters in mice with DN, including reducing 24-hour food consumption, 24-hour urine volume and 24-hour water intake and increasing body weight and serum insulin. Of note, FA treatment improved renal functional and structural damage in mice with DN. In addition, FA treatment significantly reduced the number of renal infiltrating M1 macrophages, inflammatory cytokine FA stimulation significantly reduced the increase in F4/80+CD86+ cell ratio, inflammatory factor content and p-p65/p65 protein expression induced by high glucose exposure in RAW264.7 cells. All in all, our results indicated that FA protects against kidney damage in mice with DN by inhibiting M1 macrophage polarization, and its mechanism may be related to the inhibition of nuclear factor-k-gene binding (NF-kB) signaling pathway.

Diagnostic Accuracy of the Truenat MTB Plus Assay and Comparison with the Xpert MTB/RIF Assay to Detect Tuberculosis among Hospital Outpatients in Cameroon.

The Truenat MTB Plus assay is a rapid molecular test that has been recommended by the World Health Organization since 2020 as an initial test to detect tuberculosis (TB). The WHO highlighted the need to further evaluate assay performance to inform future recommendations, including in people living with HIV and compared to the Xpert MTB/RIF assay. We conducted a prospective evaluation of the diagnostic accuracy of the Truenat assay in Cameroon, a country with a high burden of HIV/TB. Adult outpatients were recruited at four hospitals; demographic information and medical history were collected, and participants produced two sputum specimens. Truenat and Xpert testing was performed on the same specimen, and performance was compared to TB culture as the reference standard. From November 2019 to December 2020, 945 participants were enrolled and included in the analysis. Among 251 participants with culture-positive TB, the sensitivity of Truenat MTB Plus was 91% (95% confidence interval [CI], 86 to 94%), similar to Xpert (90%; 95% CI, 86 to 93%). Among 74 HIV-positive participants with culture-positive TB, the sensitivity of Truenat MTB Plus was 85% (95% CI, 75 to 92%) compared to 81% for Xpert (95% CI, 70 to 89%). Among 47 participants with smear-negative TB, the sensitivity of Truenat MTB Plus was 55% (95% CI, 40 to 70%), similar to Xpert (53%; 95% CI, 38 to 68%). The specificity of Truenat MTB Plus was 96% (95% CI, 94 to 97%) compared to 99% (95% CI, 97 to 99%) for Xpert. For TB detection compared to the reference standard of TB culture, the performance of the Truenat MTB Plus assay was similar to that of Xpert in this population, including among people living with HIV.

Impact of molecular diagnostic tests on diagnostic and treatment delays in tuberculosis: a systematic review and meta-analysis.

BACKGROUND: Countries with high TB burden have expanded access to molecular diagnostic tests. However, their impact on reducing delays in TB diagnosis and treatment has not been assessed. Our primary aim was to summarize the quantitative evidence on the impact of nucleic acid amplification tests (NAAT) on diagnostic and treatment delays compared to that of the standard of care for drug-sensitive and drug-resistant tuberculosis (DS-TB and DR-TB). METHODS: We searched MEDLINE, EMBASE, Web of Science, and the Global Health databases (from their inception to October 12, 2020) and extracted time delay data for each test. We then analysed the diagnostic and treatment initiation delay separately for DS-TB and DR-TB by comparing smear vs Xpert for DS-TB and culture drug sensitivity testing (DST) vs line probe assay (LPA) for DR-TB. We conducted random effects meta-analyses of differences of the medians to quantify the difference in diagnostic and treatment initiation delay, and we investigated heterogeneity in effect estimates based on the period the test was used in, empiric treatment rate, HIV prevalence, healthcare level, and study design. We also evaluated methodological differences in assessing time delays. RESULTS: A total of 45 studies were included in this review (DS = 26; DR = 20). We found considerable heterogeneity in the definition and reporting of time delays across the studies. For DS-TB, the use of Xpert reduced diagnostic delay by 1.79 days (95% CI - 0.27 to 3.85) and treatment initiation delay by 2.55 days (95% CI 0.54-4.56) in comparison to sputum microscopy. For DR-TB, use of LPAs reduced diagnostic delay by 40.09 days (95% CI 26.82-53.37) and treatment initiation delay by 45.32 days (95% CI 30.27-60.37) in comparison to any culture DST methods. CONCLUSIONS: Our findings indicate that the use of World Health Organization recommended diagnostics for TB reduced delays in diagnosing and initiating TB treatment. Future studies evaluating performance and impact of diagnostics should consider reporting time delay estimates based on the standardized reporting framework.

One-sample aggregate data meta-analysis of medians.

An aggregate data meta-analysis is a statistical method that pools the summary statistics of several selected studies to estimate the outcome of interest. When considering a continuous outcome, typically each study must report the same measure of the outcome variable and its spread (eg, the sample mean and its standard error). However, some studies may instead report the median along with various measures of spread. Recently, the task of incorporating medians in meta-analysis has been achieved by estimating the sample mean and its standard error from each study that reports a median in order to meta-analyze the means. In this paper, we propose two alternative approaches to meta-analyze data that instead rely on medians. We systematically compare these approaches via simulation study to each other and to methods that transform the study-specific medians and spread into sample means and their standard errors. We demonstrate that the proposed median-based approaches perform better than the transformation-based approaches, especially when applied to skewed data and data with high inter-study variance. Finally, we illustrate these approaches in a meta-analysis of patient delay in tuberculosis diagnosis.

Poultry and Fish Consumption in Relation to Total Cancer Mortality: A Meta-Analysis of Prospective Studies.

Observational studies on the association between fish and poultry intake and the risk of total cancer mortality have been reported with mixed results. Thus, we aimed to assess this association by conducting a meta-analysis of prospective studies. We performed a literature search on PubMed database through February 1, 2017 to identify relative prospective studies. A random-effects model was used to calculate the summary estimates. We identified fourteen prospective studies involving 2,378,204 participants and 96,712 cancer mortality events. Comparing the highest category of consumption with lowest category, pooled relative risk (RR) of total cancer mortality was 0.99 (95% confidence interval [CI], 0.94-1.05) for fish (10 studies) and 0.96 (95% CI, 0.93-1.00) for poultry (8 studies), respectively. However, we failed to identify any dose-response association based on a limited number of eligible studies, with a pooled RR of 0.98 (95% CI, 0.92-1.05) and 0.97 (95% CI, 0.88-1.07) for each 100 g per day increment in fish and poultry consumption, respectively. In summary, this meta-analysis indicates that consumption of either fish or poultry is not substantially associated with lower risk of total cancer mortality. The observed weak inverse association for high poultry consumption needs to be verified in additional large prospective studies.

Effects of fatty and lean fish intake on stroke risk: a meta-analysis of prospective cohort studies.

BACKGROUND: Fish intake has been postulated to reduce the risk of stroke. However, whether the beneficial effect of fish are mainly linked to fat content, as a source of omega-3 polyunsaturated fatty acids, remains unclear. We conducted a meta-analysis to compare the effect of fatty and lean fish intake on stroke risk. METHODS: We performed a literature search on four database (PubMed, Embase, Scopus, and Cochrane Library) through February 1, 2018 to identify prospective studies of fatty and lean fish in relation to stroke risk. A random-effects model was used to calculate the summary estimates. RESULTS: We identified five prospective studies, including 7 comparisons for fatty fish intake and 5 comparisons for lean fish intake. Compared with the highest category of intake with lowest category, the summary relative risk was 0.88 [95% confidence interval (CI), 0.74-1.04] for fatty fish intake and 0.81 (95% CI, 0.67-0.99) for lean fish intake. No heterogeneity across studies and publication bias were observed. CONCLUSION: Our findings demonstrated that fatty and lean fish intake has beneficial effects on stroke risk, especially lean fish intake. Additional prospective studies are necessary to confirm these observations.

An evaluation of automated chest radiography reading software for tuberculosis screening among public- and private-sector patients.

Computer-aided reading (CAR) of medical images is becoming increasingly common, but few studies exist for CAR in tuberculosis (TB). We designed a prospective study evaluating CAR for chest radiography (CXR) as a triage tool before Xpert MTB/RIF (Xpert).Consecutively enrolled adults in Dhaka, Bangladesh, with TB symptoms received CXR and Xpert. Each image was scored by CAR and graded by a radiologist. We compared CAR with the radiologist for sensitivity and specificity, area under the receiver operating characteristic curve (AUC), and calculated the potential Xpert tests saved.A total of 18 036 individuals were enrolled. TB prevalence by Xpert was 15%. The radiologist graded 49% of CXRs as abnormal, resulting in 91% sensitivity and 58% specificity. At a similar sensitivity, CAR had a lower specificity (41%), saving fewer (36%) Xpert tests. The AUC for CAR was 0.74 (95% CI 0.73-0.75). CAR performance declined with increasing age. The radiologist grading was superior across all sub-analyses.Using CAR can save Xpert tests, but the radiologist's specificity was superior. Differentiated CAR thresholds may be required for different populations. Access to, and costs of, human readers must be considered when deciding to use CAR software. More studies are needed to evaluate CAR using different screening approaches.

[Establishment of double targets of high throughput screening model for xanthine oxidase inhibitors and superoxide anion scavengers].

A double targets of high throughput screening model for xanthine oxidase inhibitors and superoxide anion scavengers was established. In the reaction system of xanthine oxidase, WST-1 works as the probe for the ultra oxygen anion generation, and product uric acid works as xanthine oxidase activity indicator. By using SpectraMax M5 continuous spectrum enzyme sign reflectoscope reflector, the changes of these indicators' concentration were observed and the influence factors of this reaction system to establish the high throughput screening model were studied. And the model is confirmed by positive drugs. In the reaction system, the final volume of reaction system is 50 µL and the concentrations of xanthine oxidase is 4 mU x mL(-1), xanthine 250 µmol x L(-1) and WST-1 100 µmol x L(-1), separately. The Z'-factor of model for xanthine oxidase inhibitors is 0.537 4, S/N is 47.519 9; the Z'-factor of model for superoxide anion scavengers is 0.507 4, S/N is 5.388 9. This model for xanthine oxidase inhibitors and superoxide anion scavengers has more common characteristics of the good stability, the fewer reagent types and quantity, the good repeatability, and so on. And it can be widely applied in high-throughput screening research.

[Pharmacokinetic comparison of two ozagrel polymorph forms in SD rats].

To enhance the quality and efficiency of ozagrel by investigating the differences between the ozagrel polymorphs in bioavailability. Solid ozagrel in different polymorph forms were orally administered to SD rats. An HPLC method was established to determinate plasma level of ozagrel. The bioavailabilities of two polymorph forms were calculated and compared. The pharmacokinetic parameters of ozagrel, were as follows: Cmax was 32.72 ± 17.04 and 34.01 ± 19.13 mg · L(-1), respectively; AUC0-t was 61.14 ± 14.76 and 85.56 ± 18.08 mg · L(-1) · h, respectively; t½ was 1.53 ± 0.51 and 4.73 ± 3.00 h, respectively. There was no significant difference in pharmacokinetic parameters between form I and II polymorphs of ozagrel while the t½ of form II is longer, which indicates that the use of form II polymorph as pharmaceutical product may prolong the effective action time in clinics. This would help the polymorph quality control in drug production.

Computer-aided detection of tuberculosis from chest radiographs in a tuberculosis prevalence survey in South Africa: external validation and modelled impacts of commercially available artificial intelligence software.

BACKGROUND: Computer-aided detection (CAD) can help identify people with active tuberculosis left undetected. However, few studies have compared the performance of commercially available CAD products for screening in high tuberculosis and high HIV settings, and there is poor understanding of threshold selection across products in different populations. We aimed to compare CAD products' performance, with further analyses on subgroup performance and threshold selection. METHODS: We evaluated 12 CAD products on a case-control sample of participants from a South African tuberculosis prevalence survey. Only those with microbiological test results were eligible. The primary outcome was comparing products' accuracy using the area under the receiver operating characteristic curve (AUC) against microbiological evidence. Threshold analyses were performed based on pre-defined criteria and across all thresholds. We conducted subgroup analyses including age, gender, HIV status, previous tuberculosis history, symptoms presence, and current smoking status. FINDINGS: Of the 774 people included, 516 were bacteriologically negative and 258 were bacteriologically positive. Diverse accuracy was noted: Lunit and Nexus had AUCs near 0·9, followed by qXR, JF CXR-2, InferRead, Xvision, and ChestEye (AUCs 0·8-0·9). XrayAME, RADIFY, and TiSepX-TB had AUC under 0·8. Thresholds varied notably across these products and different versions of the same products. Certain products (Lunit, Nexus, JF CXR-2, and qXR) maintained high sensitivity (>90%) across a wide threshold range while reducing the number of individuals requiring confirmatory diagnostic testing. All products generally performed worst in older individuals, people with previous tuberculosis, and people with HIV. Variations in thresholds, sensitivity, and specificity existed across groups and settings. INTERPRETATION: Several previously unevaluated products performed similarly to those evaluated by WHO. Thresholds differed across products and demographic subgroups. The rapid emergence of products and versions necessitates a global strategy to validate new versions and software to support CAD product and threshold selections. FUNDING: Government of Canada.

Early user experience and lessons learned using ultra-portable digital X-ray with computer-aided detection (DXR-CAD) products: A qualitative study from the perspective of healthcare providers.

BACKGROUND: Recent technological and radiological advances have renewed interest in using X-rays to screen and triage people with tuberculosis (TB). The miniaturization of digital X-ray (DXR), combined with automatic interpretation using computer-aided detection (CAD) software can extend the reach of DXR screening interventions for TB. This qualitative study assessed early implementers' experiences and lessons learned when using ultra-portable (UP) DXR systems integrated with CAD software to screen and triage TB. METHODS: Semi-structured interviews were conducted with project staff and healthcare workers at six pilot sites. Transcripts were coded and analyzed using a framework approach. The themes that emerged were subsequently organized and presented using the Consolidated Framework for Implementation Research (CFIR). RESULTS: There were 26 interviewees with varying roles: supervisory, clinicians, radiographers, and radiologists. Participants recognized the portability as the main advantage, but criticize that it involves several compromises on throughput, internet dependence, manoeuvrability, and stability, as well as suitability for patients with larger body sizes. Furthermore, compared to using hardware and software from the same supplier and without digital health information systems, complexity increases with interoperability between hardware and software, and between different electronic health information systems. Currently, there is a limited capacity to implement these technologies, especially due to the need for threshold selection, and lack of guidance on radiation protection suitable for UP DXR machines. Finally, the respondents stressed the importance of having protected means of sharing patient medical data, as well as comprehensive support and warranty plans. CONCLUSION: Study findings suggest that UP DXR with CAD was overall well received to decentralize radiological assessment for TB, however, the improved portability involved programmatic compromises. The main barriers to uptake included insufficient capacity and lack of guidance on radiation protection suitable for UP DXR.

Enhancing the reliability and accuracy of AI-enabled diagnosis via complementarity-driven deferral to clinicians.

Predictive artificial intelligence (AI) systems based on deep learning have been shown to achieve expert-level identification of diseases in multiple medical imaging settings, but can make errors in cases accurately diagnosed by clinicians and vice versa. We developed Complementarity-Driven Deferral to Clinical Workflow (CoDoC), a system that can learn to decide between the opinion of a predictive AI model and a clinical workflow. CoDoC enhances accuracy relative to clinician-only or AI-only baselines in clinical workflows that screen for breast cancer or tuberculosis (TB). For breast cancer screening, compared to double reading with arbitration in a screening program in the UK, CoDoC reduced false positives by 25% at the same false-negative rate, while achieving a 66% reduction in clinician workload. For TB triaging, compared to standalone AI and clinical workflows, CoDoC achieved a 5-15% reduction in false positives at the same false-negative rate for three of five commercially available predictive AI systems. To facilitate the deployment of CoDoC in novel futuristic clinical settings, we present results showing that CoDoC's performance gains are sustained across several axes of variation (imaging modality, clinical setting and predictive AI system) and discuss the limitations of our evaluation and where further validation would be needed. We provide an open-source implementation to encourage further research and application.

Computer-Aided Detection for Chest Radiography to Improve the Quality of Tuberculosis Diagnosis in Vietnam's District Health Facilities: An Implementation Study.

In Vietnam, chest radiography (CXR) is used to refer people for GeneXpert (Xpert) testing to diagnose tuberculosis (TB), demonstrating high yield for TB but a wide range of CXR abnormality rates. In a multi-center implementation study, computer-aided detection (CAD) was integrated into facility-based TB case finding to standardize CXR interpretation. CAD integration was guided by a programmatic framework developed for routine implementation. From April through December 2022, 24,945 CXRs from TB-vulnerable populations presenting to district health facilities were evaluated. Physicians interpreted all CXRs in parallel with CAD (qXR 3.0) software, for which the selected TB threshold score was ≥0.60. At three months, there was 47.3% concordance between physician and CAD TB-presumptive CXR results, 7.8% of individuals who received CXRs were referred for Xpert testing, and 858 people diagnosed with Xpert-confirmed TB per 100,000 CXRs. This increased at nine months to 76.1% concordant physician and CAD TB-presumptive CXRs, 9.6% referred for Xpert testing, and 2112 people with Xpert-confirmed TB per 100,000 CXRs. Our programmatic CAD-CXR framework effectively supported physicians in district facilities to improve the quality of referral for diagnostic testing and increase TB detection yield. Concordance between physician and CAD CXR results improved with training and was important to optimize Xpert testing.

Comparing different versions of computer-aided detection products when reading chest X-rays for tuberculosis.

Computer-aided detection (CAD) was recently recommended by the WHO for TB screening and triage based on several evaluations, but unlike traditional diagnostic tests, software versions are updated frequently and require constant evaluation. Since then, newer versions of two of the evaluated products have already been released. We used a case control sample of 12,890 chest X-rays to compare performance and model the programmatic effect of upgrading to newer versions of CAD4TB and qXR. We compared the area under the receiver operating characteristic curve (AUC), overall, and with data stratified by age, TB history, gender, and patient source. All versions were compared against radiologist readings and WHO's Target Product Profile (TPP) for a TB triage test. Both newer versions significantly outperformed their predecessors in terms of AUC: CAD4TB version 6 (0.823 [0.816-0.830]), version 7 (0.903 [0.897-0.908]) and qXR version 2 (0.872 [0.866-0.878]), version 3 (0.906 [0.901-0.911]). Newer versions met WHO TPP values, older versions did not. All products equalled or surpassed the human radiologist performance with improvements in triage ability in newer versions. Humans and CAD performed worse in older age groups and among those with TB history. New versions of CAD outperform their predecessors. Prior to implementation CAD should be evaluated using local data because underlying neural networks can differ significantly. An independent rapid evaluation centre is necessitated to provide implementers with performance data on new versions of CAD products as they are developed.

TB Hackathon: Development and Comparison of Five Models to Predict Subnational Tuberculosis Prevalence in Pakistan.

Pakistan's national tuberculosis control programme (NTP) is among the many programmes worldwide that value the importance of subnational tuberculosis (TB) burden estimates to support disease control efforts, but do not have reliable estimates. A hackathon was thus organised to solicit the development and comparison of several models for small area estimation of TB. The TB hackathon was launched in April 2019. Participating teams were requested to produce district-level estimates of bacteriologically positive TB prevalence among adults (over 15 years of age) for 2018. The NTP provided case-based data from their 2010-2011 TB prevalence survey, along with data relating to TB screening, testing and treatment for the period between 2010-2011 and 2018. Five teams submitted district-level TB prevalence estimates, methodological details and programming code. Although the geographical distribution of TB prevalence varied considerably across models, we identified several districts with consistently low notification-to-prevalence ratios. The hackathon highlighted the challenges of generating granular spatiotemporal TB prevalence forecasts based on a cross-sectional prevalence survey data and other data sources. Nevertheless, it provided a range of approaches to subnational disease modelling. The NTP's use and plans for these outputs shows that, limitations notwithstanding, they can be valuable for programme planning.

Tuberculosis detection from chest x-rays for triaging in a high tuberculosis-burden setting: an evaluation of five artificial intelligence algorithms.

BACKGROUND: Artificial intelligence (AI) algorithms can be trained to recognise tuberculosis-related abnormalities on chest radiographs. Various AI algorithms are available commercially, yet there is little impartial evidence on how their performance compares with each other and with radiologists. We aimed to evaluate five commercial AI algorithms for triaging tuberculosis using a large dataset that had not previously been used to train any AI algorithms. METHODS: Individuals aged 15 years or older presenting or referred to three tuberculosis screening centres in Dhaka, Bangladesh, between May 15, 2014, and Oct 4, 2016, were recruited consecutively. Every participant was verbally screened for symptoms and received a digital posterior-anterior chest x-ray and an Xpert MTB/RIF (Xpert) test. All chest x-rays were read independently by a group of three registered radiologists and five commercial AI algorithms: CAD4TB (version 7), InferRead DR (version 2), Lunit INSIGHT CXR (version 4.9.0), JF CXR-1 (version 2), and qXR (version 3). We compared the performance of the AI algorithms with each other, with the radiologists, and with the WHO's Target Product Profile (TPP) of triage tests (≥90% sensitivity and ≥70% specificity). We used a new evaluation framework that simultaneously evaluates sensitivity, proportion of Xpert tests avoided, and number needed to test to inform implementers' choice of software and selection of threshold abnormality scores. FINDINGS: Chest x-rays from 23 954 individuals were included in the analysis. All five AI algorithms significantly outperformed the radiologists. The areas under the receiver operating characteristic curve were 90·81% (95% CI 90·33-91·29) for qXR, 90·34% (89·81-90·87) for CAD4TB, 88·61% (88·03-89·20) for Lunit INSIGHT CXR, 84·90% (84·27-85·54) for InferRead DR, and 84·89% (84·26-85·53) for JF CXR-1. Only qXR (74·3% specificity [95% CI 73·3-74·9]) and CAD4TB (72·9% specificity [72·3-73·5]) met the TPP at 90% sensitivity. All five AI algorithms reduced the number of Xpert tests required by 50% while maintaining a sensitivity above 90%. All AI algorithms performed worse among older age groups (>60 years) and people with a history of tuberculosis. INTERPRETATION: AI algorithms can be highly accurate and useful triage tools for tuberculosis detection in high-burden regions, and outperform human readers. FUNDING: Government of Canada.

A new resource on artificial intelligence powered computer automated detection software products for tuberculosis programmes and implementers.

Recently, the number of artificial intelligence powered computer-aided detection (CAD) products that detect tuberculosis (TB)-related abnormalities from chest X-rays (CXR) available on the market has increased. Although CXR is a relatively effective and inexpensive method for TB screening and triaging, a shortage of skilled radiologists in many high TB-burden countries limits its use. CAD technology offers a solution to this problem. Before adopting a CAD product, TB programmes need to consider not only the diagnostic accuracy but also implementation-relevant features including operational characteristics, deployment mechanism, input and machine compatibility, output format, options for integration into the legacy system, costs, data sharing and privacy aspects, and certification. A landscaping analysis was conducted to collect this information among CAD developers known to have or soon to have a TB product. The responses were reviewed and finalized with the developers, and are published on an open-access website: www.ai4hlth.org. CAD products are constantly being improved and the site will continuously be updated to account for updates and new products. This unique online resource aims to inform the TB community about available CAD tools, their features and set-up procedures, to enable TB programmes to identify the most suitable product to incorporate in interventions.