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

Breaking the threshold: Developing multivariable models using computer-aided chest X-ray analysis for tuberculosis triage

OBJECTIVES

Computer-aided detection (CAD) software packages quantify tuberculosis (TB)-compatible chest X-ray (CXR) abnormality as continuous scores. In practice, a threshold value is selected for binary CXR classification. We assessed the diagnostic accuracy of an alternative approach to applying CAD for TB triage: incorporating CAD scores in multivariable modeling.

METHODS

We pooled individual patient data from four studies. Separately, for two commercial CAD, we used logistic regression to model microbiologically confirmed TB. Models included CAD score, study site, age, sex, human immunodeficiency virus status, and prior TB. We compared specificity at target sensitivities ≥90% between the multivariable model and the current threshold-based approach for CAD use.

RESULTS

We included 4,733/5,640 (84%) participants with complete covariate data (median age 36 years; 45% female; 22% with prior TB; 22% people living with human immunodeficiency virus). A total of 805 (17%) had TB. Multivariable models demonstrated excellent performance (areas under the receiver operating characteristic curve [95% confidence interval]: software A, 0.91 [0.90-0.93]; software B, 0.92 [0.91-0.93]). Compared with threshold scores, multivariable models increased specificity (e.g., at 90% sensitivity, threshold vs model specificity [95% confidence interval]: software A, 71% [68-74%] vs 75% [74-77%]; software B, 69% [63-75%] vs 75% [74-77%]).

CONCLUSION

Using CAD scores in multivariable models outperformed the current practice of CAD-threshold-based CXR classification for TB diagnosis.

Development and validation of a deep learning model for detecting signs of tuberculosis on chest radiographs among US-bound immigrants and refugees

Immigrants and refugees seeking admission to the United States must first undergo an overseas medical exam, overseen by the US Centers for Disease Control and Prevention (CDC), during which all persons ≥15 years old receive a chest x-ray to look for signs of tuberculosis. Although individual screening sites often implement quality control (QC) programs to ensure radiographs are interpreted correctly, the CDC does not currently have a method for conducting similar QC reviews at scale. We obtained digitized chest radiographs collected as part of the overseas immigration medical exam. Using radiographs from applicants 15 years old and older, we trained deep learning models to perform three tasks: identifying abnormal radiographs; identifying abnormal radiographs suggestive of tuberculosis; and identifying the specific findings (e.g., cavities or infiltrates) in abnormal radiographs. We then evaluated the models on both internal and external testing datasets, focusing on two classes of performance metrics: individual-level metrics, like sensitivity and specificity, and sample-level metrics, like accuracy in predicting the prevalence of abnormal radiographs. A total of 152,012 images (one image per applicant; mean applicant age 39 years) were used for model training. On our internal test dataset, our models performed well both in identifying abnormalities suggestive of TB (area under the curve [AUC] of 0.97; 95% confidence interval [CI]: 0.95, 0.98) and in estimating sample-level counts of the same (-2% absolute percentage error; 95% CIC: -8%, 6%). On the external test datasets, our models performed similarly well in identifying both generic abnormalities (AUCs ranging from 0.89 to 0.92) and those suggestive of TB (AUCs from 0.94 to 0.99). This performance was consistent across metrics, including those based on thresholded class predictions, like sensitivity, specificity, and F1 score. Strong performance relative to high-quality radiological reference standards across a variety of datasets suggests our models may make reliable tools for supporting chest radiography QC activities at CDC.

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.

Update on the diagnosis of tuberculosis

BACKGROUND

Tuberculosis (TB) remains a global public health threat, and the development of rapid and precise diagnostic tools is the key to enabling the early start of treatment, monitoring response to treatment, and preventing the spread of the disease.

OBJECTIVES

An overview of recent progress in host- and pathogen-based TB diagnostics.

SOURCES

We conducted a PubMed search of recent relevant articles and guidelines on TB screening and diagnosis.

CONTENT

An overview of currently used methods and perspectives in the following areas of TB diagnostics is provided: immune-based diagnostics, X-ray, clinical symptoms and scores, cough detection, culture of Mycobacterium tuberculosis and identifying its resistance profile using phenotypic and genotypic methods, including next-generation sequencing, sputum- and non-sputum-based molecular diagnosis of TB and monitoring of response to treatment.

IMPLICATIONS

A brief overview of the most relevant advances and changes in international guidelines regarding screening and diagnosing TB is provided in this review. It aims at reviewing all relevant areas of diagnostics, including both pathogen- and host-based methods.

Evaluation of C-Reactive Protein and Computer-Aided Analysis of Chest X-rays as Tuberculosis Triage Tests at Health Facilities in Lesotho and South Africa

BACKGROUND

To improve tuberculosis case-finding, rapid, non-sputum triage tests need to be developed according to the World Health Organization target product profile (TPP) (>90% sensitivity, >70% specificity). We prospectively evaluated and compared artificial intelligence-based, computer-aided detection software, CAD4TBv7, and C-reactive protein assay (CRP) as triage tests at health facilities in Lesotho and South Africa.

METHODS

Adults (≥18 years) presenting with ≥1 of the 4 cardinal tuberculosis symptoms were consecutively recruited between February 2021 and April 2022. After informed consent, each participant underwent a digital chest X-ray for CAD4TBv7 and a CRP test. Participants provided 1 sputum sample for Xpert MTB/RIF Ultra and Xpert MTB/RIF and 1 for liquid culture. Additionally, an expert radiologist read the chest X-rays via teleradiology. For primary analysis, a composite microbiological reference standard (ie, positive culture or Xpert Ultra) was used.

RESULTS

We enrolled 1392 participants, 48% were people with HIV and 24% had previously tuberculosis. The receiver operating characteristic curve for CAD4TBv7 and CRP showed an area under the curve of .87 (95% CI: .84-.91) and .80 (95% CI: .76-.84), respectively. At thresholds corresponding to 90% sensitivity, specificity was 68.2% (95% CI: 65.4-71.0%) and 38.2% (95% CI: 35.3-41.1%) for CAD4TBv7 and CRP, respectively. CAD4TBv7 detected tuberculosis as well as an expert radiologist. CAD4TBv7 almost met the TPP criteria for tuberculosis triage.

CONCLUSIONS

CAD4TBv7 is accurate as a triage test for patients with tuberculosis symptoms from areas with a high tuberculosis and HIV burden. The role of CRP in tuberculosis triage requires further research.

CLINICAL TRIALS REGISTRATION

Clinicaltrials.gov identifier: NCT04666311.

AI-Driven Thoracic X-ray Diagnostics: Transformative Transfer Learning for Clinical Validation in Pulmonary Radiography

Our research evaluates advanced artificial (AI) methodologies to enhance diagnostic accuracy in pulmonary radiography. Utilizing DenseNet121 and ResNet50, we analyzed 108,948 chest X-ray images from 32,717 patients and DenseNet121 achieved an area under the curve (AUC) of 94% in identifying the conditions of pneumothorax and oedema. The model's performance surpassed that of expert radiologists, though further improvements are necessary for diagnosing complex conditions such as emphysema, effusion, and hernia. Clinical validation integrating Latent Dirichlet Allocation (LDA) and Named Entity Recognition (NER) demonstrated the potential of natural language processing (NLP) in clinical workflows. The NER system achieved a precision of 92% and a recall of 88%. Sentiment analysis using DistilBERT provided a nuanced understanding of clinical notes, which is essential for refining diagnostic decisions. XGBoost and SHapley Additive exPlanations (SHAP) enhanced feature extraction and model interpretability. Local Interpretable Model-agnostic Explanations (LIME) and occlusion sensitivity analysis further enriched transparency, enabling healthcare providers to trust AI predictions. These AI techniques reduced processing times by 60% and annotation errors by 75%, setting a new benchmark for efficiency in thoracic diagnostics. The research explored the transformative potential of AI in medical imaging, advancing traditional diagnostics and accelerating medical evaluations in clinical settings.

Expanding molecular diagnostic coverage for tuberculosis by combining computer-aided chest radiography and sputum specimen pooling: a modeling study from four high-burden countries

Background

In 2022, fewer than half of persons with tuberculosis (TB) had access to molecular diagnostic tests for TB due to their high costs. Studies have found that the use of artificial intelligence (AI) software for chest X-ray (CXR) interpretation and sputum specimen pooling can each reduce the cost of testing. We modeled the combination of both strategies to estimate potential savings in consumables that could be used to expand access to molecular diagnostics.

Methods

We obtained Xpert testing and positivity data segmented into deciles by AI probability scores for TB from the community- and healthcare facility-based active case finding conducted in Bangladesh, Nigeria, Viet Nam, and Zambia. AI scores in the model were based on CAD4TB version 7 (Zambia) and qXR (all other countries). We modeled four ordinal screening and testing approaches involving AI-aided CXR interpretation to indicate individual and pooled testing. Setting a false negative rate of 5%, for each approach we calculated additional and cumulative savings over the baseline of universal Xpert testing, as well as the theoretical expansion in diagnostic coverage.

Results

In each country, the optimal screening and testing approach was to use AI to rule out testing in deciles with low AI scores and to guide pooled vs individual testing in persons with moderate and high AI scores, respectively. This approach yielded cumulative savings in Xpert tests over baseline ranging from 50.8% in Zambia to 57.5% in Nigeria and 61.5% in Bangladesh and Viet Nam. Using these savings, diagnostic coverage theoretically could be expanded by 34% to 160% across the different approaches and countries.

Conclusions

Using AI software data generated during CXR interpretation to inform a differentiated pooled testing strategy may optimize TB diagnostic test use, and could extend molecular tests to more people who need them. The optimal AI thresholds and pooled testing strategy varied across countries, which suggests that bespoke screening and testing approaches may be needed for differing populations and settings.

Supplementary Information

The online version contains supplementary material available at 10.1186/s44263-024-00081-2.

[Clinical Application of Artificial Intelligence-Based Detection Assistance Devices for Chest X-Ray Interpretation: Current Status and Practical Considerations]

Artificial intelligence (AI) technology is actively being applied for the interpretation of medical imaging, such as chest X-rays. AI-based software medical devices, which automatically detect various types of abnormal findings in chest X-ray images to assist physicians in their interpretation, are actively being commercialized and clinically implemented in Korea. Several important issues need to be considered for AI-based detection assistant tools to be applied in clinical practice: the evaluation of performance and efficacy prior to implementation; the determination of the target application, range, and method of delivering results; and monitoring after implementation and legal liability issues. Appropriate decision making regarding these devices based on the situation in each institution is necessary. Radiologists must be engaged as medical assessment experts using the software for these devices as well as in medical image interpretation to ensure the safe and efficient implementation and operation of AI-based detection assistant tools.

What is New in the Diagnosis of Childhood Tuberculosis?

The fact that almost half of the 1 million cases of childhood tuberculosis (TB) globally remain undiagnosed jeopardizes the TB elimination goal. Fortunately, there are new advances in this field which have the potential to bridge this diagnostic gap. Advances in imaging include computer assisted interpretation of chest X-rays (CXRs), point of care ultrasound (POCUS) and faster and superior computed tomography/ magnetic resonance imaging (CT/ MRI) protocols. The urine lipoarabinomannan test has proved to be a good point of care test for diagnosing TB in Human immunodeficiency virus (HIV) infected children. Stool and nasopharyngeal aspirates are emerging as acceptable alternatives for gastric lavage and induced sputum for diagnosing intrathoracic tuberculosis. Xpert MTB/RIF Ultra has improved sensitivity compared to Xpert MTB/RIF for diagnosing both pulmonary/ extrapulmonary TB. Xpert XDR is another commercially available accurate point of care test for detecting resistance to drugs other than rifampicin in smear positive samples. Other molecular methods including new line probe assays, pyrosequencing, whole genome sequencing, and targeted next generation sequencing are extremely promising but not available commercially at present. The C-Tb skin test is an acceptable alternative to the tuberculin skin test and interferon gamma release assays for diagnosis of latent infection. There is an urgent need to incorporate some of these advances in the existing diagnostic algorithms of childhood TB.

A deep learning-based algorithm for pulmonary tuberculosis detection in chest radiography

In tuberculosis (TB), chest radiography (CXR) patterns are highly variable, mimicking pneumonia and many other diseases. This study aims to evaluate the efficacy of Google teachable machine, a deep neural network-based image classification tool, to develop algorithm for predicting TB probability of CXRs. The training dataset included 348 TB CXRs and 3806 normal CXRs for training TB detection. We also collected 1150 abnormal CXRs and 627 normal CXRs for training abnormality detection. For external validation, we collected 250 CXRs from our hospital. We also compared the accuracy of the algorithm to five pulmonologists and radiological reports. In external validation, the AI algorithm showed areas under the curve (AUC) of 0.951 and 0.975 in validation dataset 1 and 2. The accuracy of the pulmonologists on validation dataset 2 showed AUC range of 0.936-0.995. When abnormal CXRs other than TB were added, AUC decreased in both human readers (0.843-0.888) and AI algorithm (0.828). When combine human readers with AI algorithm, the AUC further increased to 0.862-0.885. The TB CXR AI algorithm developed by using Google teachable machine in this study is effective, with the accuracy close to experienced clinical physicians, and may be helpful for detecting tuberculosis by CXR.

An independent, multi-country head-to-head accuracy comparison of automated chest x-ray algorithms for the triage of pulmonary tuberculosis

Background

Computer-aided detection (CAD) algorithms for automated chest X-ray (CXR) reading have been endorsed by the World Health Organization for tuberculosis (TB) triage, but independent, multi-country assessment and comparison of current products are needed to guide implementation.

Methods

We conducted a head-to-head evaluation of five CAD algorithms for TB triage across seven countries. We included CXRs from adults who presented to outpatient facilities with at least two weeks of cough in India, Madagascar, the Philippines, South Africa, Tanzania, Uganda, and Vietnam. The participants completed a standard evaluation for pulmonary TB, including sputum collection for Xpert MTB/RIF Ultra and culture. Against a microbiological reference standard, we calculated and compared the accuracy overall, by country and key groups for five CAD algorithms: CAD4TB (Delft Imaging), INSIGHT CXR (Lunit), DrAid (Vinbrain), Genki (Deeptek), and qXR (qure.AI). We determined the area under the ROC curve (AUC) and if any CAD product could achieve the minimum target accuracy for a TB triage test (≥90% sensitivity and ≥70% specificity). We then applied country- and population-specific thresholds and recalculated accuracy to assess any improvement in performance.

Results

Of 3,927 individuals included, the median age was 41 years (IQR 29-54), 12.9% were people living with HIV (PLWH), 8.2% living with diabetes, and 21.2% had a prior history of TB. The overall AUC ranged from 0.774-0.819, and specificity ranged from 64.8-73.8% at 90% sensitivity. CAD4TB had the highest overall accuracy (73.8% specific, 95% CI 72.2-75.4, at 90% sensitivity), although qXR and INSIGHT CXR also achieved the target 70% specificity. There was heterogeneity in accuracy by country, and females and PLWH had lower sensitivity while males and people with a history of TB had lower specificity. The performance remained stable regardless of diabetes status. When country- and population-specific thresholds were applied, at least one CAD product could achieve or approach the target accuracy for each country and sub-group, except for PLWH and those with a history of TB.

Conclusions

Multiple CAD algorithms can achieve or exceed the minimum target accuracy for a TB triage test, with improvement when using setting- or population-specific thresholds. Further efforts are needed to integrate CAD into routine TB case detection programs in high-burden communities.

Automated Pulmonary Tuberculosis Severity Assessment on Chest X-rays

According to the 2022 World Health Organization's Global Tuberculosis (TB) report, an estimated 10.6 million people fell ill with TB, and 1.6 million died from the disease in 2021. In addition, 2021 saw a reversal of a decades-long trend of declining TB infections and deaths, with an estimated increase of 4.5% in the number of people who fell ill with TB compared to 2020, and an estimated yearly increase of 450,000 cases of drug resistant TB. Estimating the severity of pulmonary TB using frontal chest X-rays (CXR) can enable better resource allocation in resource constrained settings and monitoring of treatment response, enabling prompt treatment modifications if disease severity does not decrease over time. The Timika score is a clinically used TB severity score based on a CXR reading. This work proposes and evaluates three deep learning-based approaches for predicting the Timika score with varying levels of explainability. The first approach uses two deep learning-based models, one to explicitly detect lesion regions using YOLOV5n and another to predict the presence of cavitation using DenseNet121, which are then utilized in score calculation. The second approach uses a DenseNet121-based regression model to directly predict the affected lung percentage and another to predict cavitation presence using a DenseNet121-based classification model. Finally, the third approach directly predicts the Timika score using a DenseNet121-based regression model. The best performance is achieved by the second approach with a mean absolute error of 13-14% and a Pearson correlation of 0.7-0.84 using three held-out datasets for evaluating generalization.

Deep Learning-based Diagnosis of Pulmonary Tuberculosis on Chest X-ray in the Emergency Department: A Retrospective Study

Prompt and correct detection of pulmonary tuberculosis (PTB) is critical in preventing its spread. We aimed to develop a deep learning-based algorithm for detecting PTB on chest X-ray (CXRs) in the emergency department. This retrospective study included 3498 CXRs acquired from the National Taiwan University Hospital (NTUH). The images were chronologically split into a training dataset, NTUH-1519 (images acquired during the years 2015 to 2019; n = 2144), and a testing dataset, NTUH-20 (images acquired during the year 2020; n = 1354). Public databases, including the NIH ChestX-ray14 dataset (model training; 112,120 images), Montgomery County (model testing; 138 images), and Shenzhen (model testing; 662 images), were also used in model development. EfficientNetV2 was the basic architecture of the algorithm. Images from ChestX-ray14 were employed for pseudo-labelling to perform semi-supervised learning. The algorithm demonstrated excellent performance in detecting PTB (area under the receiver operating characteristic curve [AUC] 0.878, 95% confidence interval [CI] 0.854-0.900) in NTUH-20. The algorithm showed significantly better performance in posterior-anterior (PA) CXR (AUC 0.940, 95% CI 0.912-0.965, p-value < 0.001) compared with anterior-posterior (AUC 0.782, 95% CI 0.644-0.897) or portable anterior-posterior (AUC 0.869, 95% CI 0.814-0.918) CXR. The algorithm accurately detected cases of bacteriologically confirmed PTB (AUC 0.854, 95% CI 0.823-0.883). Finally, the algorithm tested favourably in Montgomery County (AUC 0.838, 95% CI 0.765-0.904) and Shenzhen (AUC 0.806, 95% CI 0.771-0.839). A deep learning-based algorithm could detect PTB on CXR with excellent performance, which may help shorten the interval between detection and airborne isolation for patients with PTB.

COVID-19 and Tuberculosis: Mathematical Modeling of Infection Spread Taking into Account Reduced Screening

The COVID-19 pandemic resulted in the cessation of many tuberculosis (TB) support programs and reduced screening coverage for TB worldwide. We propose a model that demonstrates, among other things, how undetected cases of TB affect the number of future M. tuberculosis (M. tb) infections. The analysis of official statistics on the incidence of TB, preventive examination coverage of the population, and the number of patients with bacterial excretion of M. tb in the Russian Federation from 2008 to 2021 is carried out. The desired model can be obtained due to the fluctuation of these indicators in 2020, when the COVID-19 pandemic caused a dramatic reduction in TB interventions. Statistical analysis is carried out using R v.4.2.1. The resulting model describes the dependence of the detected incidence and prevalence of TB with bacterial excretion in the current year on the prevalence of TB with bacterial excretion in the previous year and on the coverage of preventive examinations in the current and previous years. The adjusted coefficient of model determination (adjusted R-squared) is 0.9969, indicating that the model contains almost no random component. It clearly shows that TB cases missed due to low screening coverage and left uncontrolled will lead to a significant increase in the number of new infections in the future. We may conclude that the obtained results clearly demonstrate the need for mass screening of the population in the context of the spread of TB infection, which makes it possible to timely identify patients with TB with bacterial excretion.

Preliminary Evidence of the Use of Generative AI in Health Care Clinical Services: Systematic Narrative Review

BACKGROUND

Generative artificial intelligence tools and applications (GenAI) are being increasingly used in health care. Physicians, specialists, and other providers have started primarily using GenAI as an aid or tool to gather knowledge, provide information, train, or generate suggestive dialogue between physicians and patients or between physicians and patients' families or friends. However, unless the use of GenAI is oriented to be helpful in clinical service encounters that can improve the accuracy of diagnosis, treatment, and patient outcomes, the expected potential will not be achieved. As adoption continues, it is essential to validate the effectiveness of the infusion of GenAI as an intelligent technology in service encounters to understand the gap in actual clinical service use of GenAI.

OBJECTIVE

This study synthesizes preliminary evidence on how GenAI assists, guides, and automates clinical service rendering and encounters in health care The review scope was limited to articles published in peer-reviewed medical journals.

METHODS

We screened and selected 0.38% (161/42,459) of articles published between January 1, 2020, and May 31, 2023, identified from PubMed. We followed the protocols outlined in the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines to select highly relevant studies with at least 1 element on clinical use, evaluation, and validation to provide evidence of GenAI use in clinical services. The articles were classified based on their relevance to clinical service functions or activities using the descriptive and analytical information presented in the articles.

RESULTS

Of 161 articles, 141 (87.6%) reported using GenAI to assist services through knowledge access, collation, and filtering. GenAI was used for disease detection (19/161, 11.8%), diagnosis (14/161, 8.7%), and screening processes (12/161, 7.5%) in the areas of radiology (17/161, 10.6%), cardiology (12/161, 7.5%), gastrointestinal medicine (4/161, 2.5%), and diabetes (6/161, 3.7%). The literature synthesis in this study suggests that GenAI is mainly used for diagnostic processes, improvement of diagnosis accuracy, and screening and diagnostic purposes using knowledge access. Although this solves the problem of knowledge access and may improve diagnostic accuracy, it is oriented toward higher value creation in health care.

CONCLUSIONS

GenAI informs rather than assisting or automating clinical service functions in health care. There is potential in clinical service, but it has yet to be actualized for GenAI. More clinical service-level evidence that GenAI is used to streamline some functions or provides more automated help than only information retrieval is needed. To transform health care as purported, more studies related to GenAI applications must automate and guide human-performed services and keep up with the optimism that forward-thinking health care organizations will take advantage of GenAI.

Comparison of the Discrimination Performance of AI Scoring and the Brixia Score in Predicting COVID-19 Severity on Chest X-Ray Imaging: Diagnostic Accuracy Study

BACKGROUND

The artificial intelligence (AI) analysis of chest x-rays can increase the precision of binary COVID-19 diagnosis. However, it is unknown if AI-based chest x-rays can predict who will develop severe COVID-19, especially in low- and middle-income countries.

OBJECTIVE

The study aims to compare the performance of human radiologist Brixia scores versus 2 AI scoring systems in predicting the severity of COVID-19 pneumonia.

METHODS

We performed a cross-sectional study of 300 patients suspected with and with confirmed COVID-19 infection in Jakarta, Indonesia. A total of 2 AI scores were generated using CAD4COVID x-ray software.

RESULTS

The AI probability score had slightly lower discrimination (area under the curve [AUC] 0.787, 95% CI 0.722-0.852). The AI score for the affected lung area (AUC 0.857, 95% CI 0.809-0.905) was almost as good as the human Brixia score (AUC 0.863, 95% CI 0.818-0.908).

CONCLUSIONS

The AI score for the affected lung area and the human radiologist Brixia score had similar and good discrimination performance in predicting COVID-19 severity. Our study demonstrated that using AI-based diagnostic tools is possible, even in low-resource settings. However, before it is widely adopted in daily practice, more studies with a larger scale and that are prospective in nature are needed to confirm our findings.

AI for Detection of Tuberculosis: Implications for Global Health

Tuberculosis, which primarily affects developing countries, remains a significant global health concern. Since the 2010s, the role of chest radiography has expanded in tuberculosis triage and screening beyond its traditional complementary role in the diagnosis of tuberculosis. Computer-aided diagnosis (CAD) systems for tuberculosis detection on chest radiographs have recently made substantial progress in diagnostic performance, thanks to deep learning technologies. The current performance of CAD systems for tuberculosis has approximated that of human experts, presenting a potential solution to the shortage of human readers to interpret chest radiographs in low- or middle-income, high-tuberculosis-burden countries. This article provides a critical appraisal of developmental process reporting in extant CAD software for tuberculosis, based on the Checklist for Artificial Intelligence in Medical Imaging. It also explores several considerations to scale up CAD solutions, encompassing manufacturer-independent CAD validation, economic and political aspects, and ethical concerns, as well as the potential for broadening radiography-based diagnosis to other nontuberculosis diseases. Collectively, CAD for tuberculosis will emerge as a representative deep learning application, catalyzing advances in global health and health equity. Keywords: Computer-aided Diagnosis (CAD), Conventional Radiography, Thorax, Lung, Machine Learning Supplemental material is available for this article. © RSNA, 2024.

Deep learning for precise diagnosis and subtype triage of drug-resistant tuberculosis on chest computed tomography

Deep learning, transforming input data into target prediction through intricate network structures, has inspired novel exploration in automated diagnosis based on medical images. The distinct morphological characteristics of chest abnormalities between drug-resistant tuberculosis (DR-TB) and drug-sensitive tuberculosis (DS-TB) on chest computed tomography (CT) are of potential value in differential diagnosis, which is challenging in the clinic. Hence, based on 1176 chest CT volumes from the equal number of patients with tuberculosis (TB), we presented a Deep learning-based system for TB drug resistance identification and subtype classification (DeepTB), which could automatically diagnose DR-TB and classify crucial subtypes, including rifampicin-resistant tuberculosis, multidrug-resistant tuberculosis, and extensively drug-resistant tuberculosis. Moreover, chest lesions were manually annotated to endow the model with robust power to assist radiologists in image interpretation and the Circos revealed the relationship between chest abnormalities and specific types of DR-TB. Finally, DeepTB achieved an area under the curve (AUC) up to 0.930 for thoracic abnormality detection and 0.943 for DR-TB diagnosis. Notably, the system demonstrated instructive value in DR-TB subtype classification with AUCs ranging from 0.880 to 0.928. Meanwhile, class activation maps were generated to express a human-understandable visual concept. Together, showing a prominent performance, DeepTB would be impactful in clinical decision-making for DR-TB.

Analysis of Vocal Signatures of COVID-19 in Cough Sounds: A Newer Diagnostic Approach Using Artificial Intelligence

BACKGROUND

Artificial intelligence (AI) based models are explored increasingly in the medical field. The highly contagious pandemic of coronavirus disease 2019 (COVID-19) affected the world and availability of diagnostic tools high resolution computed tomography (HRCT) and/or real-time reverse transcriptase-polymerase chain reaction (RTPCR) was very limited, costly and time consuming. Therefore, the use of AI in COVID-19 for diagnosis using cough sounds can be efficacious and cost effective for screening in clinic or hospital and help in early diagnosis and further management of patients.

OBJECTIVES

To develop an accurate and fast voice-processing AI software to determine voice-based signatures in discriminating COVID-19 and non-COVID-19 cough sounds for screening of COVID-19.

METHODOLOGY

A prospective study involving 117 patients was performed based on online and/or offline voice data collection of cough sounds of COVID-19 patients in isolation ward of a tertiary care teaching hospital and non-COVID-19 participants using a smart phone. A website-based AI software was developed to identify the cough sounds as COVID-19 or non-COVID-19. The data were divided into three segments including training set, validation set and test set. A pre-processing algorithm was utilized and combined with Short Time Fourier Transform feature representation and Logistic regression model. A precise software was used to identify vocal signatures and K-fold cross validation was carried out.

RESULT

A total of 117 audio recordings of cough sounds were collected through the developed website after inclusion-exclusion criteria out of which 52 have been marked belonging to COVID-19 positive, while 65 were marked as COVID-19 negative/unsure /never had COVID-19, which were assumed to be COVID-19 negative based on RT-PCR test results. The mean and standard error values for the accuracies attained at the end of each experiment in training, validation and testing set were found to be 67.34%±0.22, 58.57%±1.11 and 64.60%±1.79 respectively. The weight values were found to be positive which were contributing towards predicting the samples as COVID-19 positive with large spikes around 7.5 kHz, 7.8 kHz, 8.6 kHz and 11 kHz which can be used for classification.

CONCLUSION

The proposed AI based approach can be a helpful screening tool for COVID-19 using vocal sounds of cough. It can help the health system by reducing the cost burden and improving overall diagnosis and management of the disease.

Accuracy of digital chest x-ray analysis with artificial intelligence software as a triage and screening tool in hospitalized patients being evaluated for tuberculosis in Lima, Peru

Tuberculosis (TB) transmission in healthcare facilities is common in high-incidence countries. Yet, the optimal approach for identifying inpatients who may have TB is unclear. We evaluated the diagnostic accuracy of qXR (Qure.ai, India) computer-aided detection (CAD) software versions 3.0 and 4.0 (v3 and v4) as a triage and screening tool within the FAST (Find cases Actively, Separate safely, and Treat effectively) transmission control strategy. We prospectively enrolled two cohorts of patients admitted to a tertiary hospital in Lima, Peru: one group had cough or TB risk factors (triage) and the other did not report cough or TB risk factors (screening). We evaluated the sensitivity and specificity of qXR for the diagnosis of pulmonary TB using culture and Xpert as primary and secondary reference standards, including stratified analyses based on risk factors. In the triage cohort (n = 387), qXR v4 sensitivity was 0.91 (59/65, 95% CI 0.81-0.97) and specificity was 0.32 (103/322, 95% CI 0.27-0.37) using culture as reference standard. There was no difference in the area under the receiver-operating-characteristic curve (AUC) between qXR v3 and qXR v4 with either a culture or Xpert reference standard. In the screening cohort (n = 191), only one patient had a positive Xpert result, but specificity in this cohort was high (>90%). A high prevalence of radiographic lung abnormalities, most notably opacities (81%), consolidation (62%), or nodules (58%), was detected by qXR on digital CXR images from the triage cohort. qXR had high sensitivity but low specificity as a triage in hospitalized patients with cough or TB risk factors. Screening patients without cough or risk factors in this setting had a low diagnostic yield. These findings further support the need for population and setting-specific thresholds for CAD programs.

Post-treatment Radiographic Severity and Mortality in Mycobacterium avium Complex Pulmonary Disease

Rationale: Imaging studies are widely performed when treating Mycobacterium avium complex pulmonary disease (MAC-PD); however, the clinical significance of post-treatment radiographic change is unknown. Objectives: To determine whether a deep neural network trained with pulmonary tuberculosis could adequately score the radiographic severity of MAC-PD and then to examine relationships between post-treatment radiographic severity and its change from baseline and long-term prognosis. Methods: We retrospectively collected chest radiographs of adult patients with MAC-PD treated for ⩾6 months at baseline and at 3, 6, 9, and 12 months of treatment. We correlated the radiographic severity score generated by a deep neural network with visual and clinical severity as determined by radiologists and mycobacterial culture status, respectively. The associations between the score, improvement from baseline, and mortality were analyzed using Cox proportional hazards regression. Results: In total, 342 and 120 patients were included in the derivation and validation cohorts, respectively. The network's severity score correlated with radiologists' grading (Spearman coefficient, 0.40) and mycobacterial culture results (odds ratio, 1.02; 95% confidence interval [CI], 1.0-1.05). A significant decreasing trend in the severity score was observed over time (P < 0.001). A higher score at 12 months of treatment was independently associated with higher mortality (adjusted hazard ratio, 1.07; 95% CI, 1.03-1.10). Improvements in radiographic scores from baseline were associated with reduced mortality, regardless of culture conversion (adjusted hazard ratio, 0.42; 95% CI, 0.22-0.80). These findings were replicated in the validation cohort. Conclusions: Post-treatment radiographic severity and improvement from baseline in patients with MAC-PD were associated with long-term survival.

Diagnostic Accuracy of Computer-Aided Detection During Active Case Finding for Pulmonary Tuberculosis in Africa: A Systematic Review and Meta-analysis

Background

Computer-aided detection (CAD) may be a useful screening tool for tuberculosis (TB). However, there are limited data about its utility in active case finding (ACF) in a community-based setting, and particularly in an HIV-endemic setting where performance may be compromised.

Methods

We performed a systematic review and evaluated articles published between January 2012 and February 2023 that included CAD as a screening tool to detect pulmonary TB against a microbiological reference standard (sputum culture and/or nucleic acid amplification test [NAAT]). We collected and summarized data on study characteristics and diagnostic accuracy measures. Two reviewers independently extracted data and assessed methodological quality against Quality Assessment of Diagnostic Accuracy Studies-2 criteria. Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Diagnostic Test Accuracy Studies (PRISMA-DTA) guidelines were followed.

Results

Of 1748 articles reviewed, 5 met with the eligibility criteria and were included in this review. A meta-analysis revealed pooled sensitivity of 0.87 (95% CI, 0.78-0.96) and specificity of 0.74 (95% CI, 0.55-0.93), just below the World Health Organization (WHO)-recommended target product profile (TPP) for a screening test (sensitivity ≥0.90 and specificity ≥0.70). We found a high risk of bias and applicability concerns across all studies. Subgroup analyses, including the impact of HIV and previous TB, were not possible due to the nature of the reporting within the included studies.

Conclusions

This review provides evidence, specifically in the context of ACF, for CAD as a potentially useful and cost-effective screening tool for TB in a resource-poor HIV-endemic African setting. However, given methodological concerns, caution is required with regards to applicability and generalizability.

[The contribution of artificial intelligence (AI) subsequent to the processing of thoracic imaging]

The contribution of artificial intelligence (AI) to medical imaging is currently the object of widespread experimentation. The development of deep learning (DL) methods, particularly convolution neural networks (CNNs), has led to performance gains often superior to those achieved by conventional methods such as machine learning. Radiomics is an approach aimed at extracting quantitative data not accessible to the human eye from images expressing a disease. The data subsequently feed machine learning models and produce diagnostic or prognostic probabilities. As for the multiple applications of AI methods in thoracic imaging, they are undergoing evaluation. Chest radiography is a practically ideal field for the development of DL algorithms able to automatically interpret X-rays. Current algorithms can detect up to 14 different abnormalities present either in isolation or in combination. Chest CT is another area offering numerous AI applications. Various algorithms have been specifically formed and validated for the detection and characterization of pulmonary nodules and pulmonary embolism, as well as segmentation and quantitative analysis of the extent of diffuse lung diseases (emphysema, infectious pneumonias, interstitial lung disease). In addition, the analysis of medical images can be associated with clinical, biological, and functional data (multi-omics analysis), the objective being to construct predictive approaches regarding disease prognosis and response to treatment.

Application of multi-gene genetic programming to the prognosis prediction of COVID-19 using routine hematological variables

Identifying patients who may develop severe COVID-19 has been of interest to clinical physicians since it facilitates personalized treatment and optimizes the allocation of medical resources. In this study, multi-gene genetic programming (MGGP), as an advanced artificial intelligence (AI) tool, was used to determine the importance of laboratory predictors in the prognosis of COVID-19 patients. The present retrospective study was conducted on 1455 patients with COVID-19 (727 males and 728 females), who were admitted to Allameh Behlool Gonabadi Hospital, Gonabad, Iran in 2020-2021. For each patient, the demographic characteristics, common laboratory tests at the time of admission, duration of hospitalization, admission to the intensive care unit (ICU), and mortality were collected through the electronic information system of the hospital. Then, the data were normalized and randomly divided into training and test data. Furthermore, mathematical prediction models were developed by MGGP for each gender. Finally, a sensitivity analysis was performed to determine the significance of input parameters on the COVID-19 prognosis. Based on the achieved results, MGGP is able to predict the mortality of COVID-19 patients with an accuracy of 60-92%, the duration of hospital stay with an accuracy of 53-65%, and admission to the ICU with an accuracy of 76-91%, using common hematological tests at the time of admission. Also, sensitivity analysis indicated that blood urea nitrogen (BUN) and aspartate aminotransferase (AST) play key roles in the prognosis of COVID-19 patients. AI techniques, such as MGGP, can be used in the triage and prognosis prediction of COVID-19 patients. In addition, due to the sensitivity of BUN and AST in the estimation models, further studies on the role of the mentioned parameters in the pathophysiology of COVID-19 are recommended.

Applicability of artificial intelligence-based computer-aided detection (AI-CAD) for pulmonary tuberculosis to community-based active case finding

BACKGROUND

Artificial intelligence-based computer-aided detection (AI-CAD) for tuberculosis (TB) has become commercially available and several studies have been conducted to evaluate the performance of AI-CAD for pulmonary tuberculosis (TB) in clinical settings. However, little is known about its applicability to community-based active case-finding (ACF) for TB.

METHODS

We analysed an anonymized data set obtained from a community-based ACF in Cambodia, targeting persons aged 55 years or over, persons with any TB symptoms, such as chronic cough, and persons at risk of TB, including household contacts. All of the participants in the ACF were screened by chest radiography (CXR) by Cambodian doctors, followed by Xpert test when they were eligible for sputum examination. Interpretation by an experienced chest physician and abnormality scoring by a newly developed AI-CAD were retrospectively conducted for the CXR images. With a reference of Xpert-positive TB or human interpretations, receiver operating characteristic (ROC) curves were drawn to evaluate the AI-CAD performance by area under the ROC curve (AUROC). In addition, its applicability to community-based ACFs in Cambodia was examined.

RESULTS

TB scores of the AI-CAD were significantly associated with the CXR classifications as indicated by the severity of TB disease, and its AUROC as the bacteriological reference was 0.86 (95% confidence interval 0.83-0.89). Using a threshold for triage purposes, the human reading and bacteriological examination needed fell to 21% and 15%, respectively, detecting 95% of Xpert-positive TB in ACF. For screening purposes, we could detect 98% of Xpert-positive TB cases.

CONCLUSIONS

AI-CAD is applicable to community-based ACF in high TB burden settings, where experienced human readers for CXR images are scarce. The use of AI-CAD in developing countries has the potential to expand CXR screening in community-based ACFs, with a substantial decrease in the workload on human readers and laboratory labour. Further studies are needed to generalize the results to other countries by increasing the sample size and comparing the AI-CAD performance with that of more human readers.

Computer-aided detection thresholds for digital chest radiography interpretation in tuberculosis diagnostic algorithms

Objectives

Use of computer-aided detection (CAD) software is recommended to improve tuberculosis screening and triage, but threshold determination is challenging if reference testing has not been performed in all individuals. We aimed to determine such thresholds through secondary analysis of the 2019 Lesotho national tuberculosis prevalence survey.

Methods

Symptom screening and chest radiographs were performed in participants aged ≥15 years; those symptomatic or with abnormal chest radiographs provided samples for Xpert MTB/RIF and culture testing. Chest radiographs were processed using CAD4TB version 7. We used six methodological approaches to deal with participants who did not have bacteriological test results to estimate pulmonary tuberculosis prevalence and assess diagnostic accuracy.

Results

Among 17 070 participants, 5214 (31%) had their tuberculosis status determined; 142 had tuberculosis. Prevalence estimates varied between methodological approaches (0.83-2.72%). Using multiple imputation to estimate tuberculosis status for those eligible but not tested, and assuming those not eligible for testing were negative, a CAD4TBv7 threshold of 13 had a sensitivity of 89.7% (95% CI 84.6-94.8) and a specificity of 74.2% (73.6-74.9), close to World Health Organization (WHO) target product profile criteria. Assuming all those not tested were negative produced similar results.

Conclusions

This is the first study to evaluate CAD4TB in a community screening context employing a range of approaches to account for unknown tuberculosis status. The assumption that those not tested are negative - regardless of testing eligibility status - was robust. As threshold determination must be context specific, our analytically straightforward approach should be adopted to leverage prevalence surveys for CAD threshold determination in other settings with a comparable proportion of eligible but not tested participants.

Accuracy of digital chest x-ray analysis with artificial intelligence software as a triage and screening tool in hospitalized patients being evaluated for tuberculosis in Lima, Peru

Introduction

Tuberculosis (TB) transmission in healthcare facilities is common in high-incidence countries. Yet, the optimal approach for identifying inpatients who may have TB is unclear. We evaluated the diagnostic accuracy of qXR (Qure.ai, India) computer-aided detection (CAD) software versions 3.0 and 4.0 (v3 and v4) as a triage and screening tool within the FAST (Find cases Actively, Separate safely, and Treat effectively) transmission control strategy.

Methods

We prospectively enrolled two cohorts of patients admitted to a tertiary hospital in Lima, Peru: one group had cough or TB risk factors (triage) and the other did not report cough or TB risk factors (screening). We evaluated the sensitivity and specificity of qXR for the diagnosis of pulmonary TB using culture and Xpert as primary and secondary reference standards, including stratified analyses based on risk factors.

Results

In the triage cohort (n=387), qXR v4 sensitivity was 0.91 (59/65, 95% CI 0.81-0.97) and specificity was 0.32 (103/322, 95% CI 0.27-0.37) using culture as reference standard. There was no difference in the area under the receiver-operating-characteristic curve (AUC) between qXR v3 and qXR v4 with either a culture or Xpert reference standard. In the screening cohort (n=191), only one patient had a positive Xpert result, but specificity in this cohort was high (>90%). A high prevalence of radiographic lung abnormalities, most notably opacities (81%), consolidation (62%), or nodules (58%), was detected by qXR on digital CXR images from the triage cohort.

Conclusions

qXR had high sensitivity but low specificity as a triage in hospitalized patients with cough or TB risk factors. Screening patients without cough or risk factors in this setting had a low diagnostic yield. These findings further support the need for population and setting-specific thresholds for CAD programs.

Implementing a chest X-ray artificial intelligence tool to enhance tuberculosis screening in India: Lessons learned

Artificial Intelligence (AI) based chest X-ray (CXR) screening for tuberculosis (TB) is becoming increasingly popular. Still, deploying such AI tools can be challenging due to multiple real-life barriers like software installation, workflow integration, network connectivity constraints, limited human resources available to interpret findings, etc. To understand these challenges, PATH implemented a TB REACH active case-finding program in a resource-limited setting of Nagpur in India, where an AI software device (qXR) intended for TB screening using CXR images was used. Eight private CXR laboratories that fulfilled prerequisites for AI software installation were engaged for this program. Key lessons about operational feasibility and accessibility, along with the strategies adopted to overcome these challenges, were learned during this program. This program also helped to screen 10,481 presumptive TB individuals using informal providers based on clinical history. Among them, 2,303 individuals were flagged as presumptive for TB by a radiologist or by AI based on their CXR interpretation. Approximately 15.8% increase in overall TB yield could be attributed to the presence of AI alone because these additional cases were not deemed presumptive for TB by radiologists, but AI was able to identify them. Successful implementation of AI tools like qXR in resource-limited settings in India will require solving real-life implementation challenges for seamless deployment and workflow integration.

Effect of multimodal diagnostic approach using deep learning-based automated detection algorithm for active pulmonary tuberculosis

In this study, we developed a model to predict culture test results for pulmonary tuberculosis (PTB) with a customized multimodal approach and evaluated its performance in different clinical settings. Moreover, we investigated potential performance improvements by combining this approach with deep learning-based automated detection algorithms (DLADs). This retrospective observational study enrolled patients over 18 years of age who consecutively visited the level 1 emergency department and underwent chest radiograph and sputum testing. The primary endpoint was positive sputum culture for PTB. We compared the performance of the diagnostic models by replacing radiologists' interpretations of chest radiographs with screening scores calculated through DLAD. The optimal diagnostic model had an area under the receiver operating characteristic curve of 0.924 (95% CI 0.871-0.976) and an area under precision recall curve of 0.403 (95% CI 0.195-0.580) while maintaining a specificity of 81.4% when sensitivity was fixed at 90%. Multicomponent models showed improved performance for detecting PTB when chest radiography interpretation was replaced by DLAD. Multicomponent diagnostic models with DLAD customized for different clinical settings are more practical than traditional methods for detecting patients with PTB. This novel diagnostic approach may help prevent the spread of PTB and optimize healthcare resource utilization in resource-limited clinical settings.

Using Artificial Intelligence to Stratify Normal versus Abnormal Chest X-rays: External Validation of a Deep Learning Algorithm at East Kent Hospitals University NHS Foundation Trust

Background: The chest radiograph (CXR) is the most frequently performed radiological examination worldwide. The increasing volume of CXRs performed in hospitals causes reporting backlogs and increased waiting times for patients, potentially compromising timely clinical intervention and patient safety. Implementing computer-aided detection (CAD) artificial intelligence (AI) algorithms capable of accurate and rapid CXR reporting could help address such limitations. A novel use for AI reporting is the classification of CXRs as 'abnormal' or 'normal'. This classification could help optimize resource allocation and aid radiologists in managing their time efficiently. Methods: qXR is a CE-marked computer-aided detection (CAD) software trained on over 4.4 million CXRs. In this retrospective cross-sectional pre-deployment study, we evaluated the performance of qXR in stratifying normal and abnormal CXRs. We analyzed 1040 CXRs from various referral sources, including general practices (GP), Accident and Emergency (A&E) departments, and inpatient (IP) and outpatient (OP) settings at East Kent Hospitals University NHS Foundation Trust. The ground truth for the CXRs was established by assessing the agreement between two senior radiologists. Results: The CAD software had a sensitivity of 99.7% and a specificity of 67.4%. The sub-group analysis showed no statistically significant difference in performance across healthcare settings, age, gender, and X-ray manufacturer. Conclusions: The study showed that qXR can accurately stratify CXRs as normal versus abnormal, potentially reducing reporting backlogs and resulting in early patient intervention, which may result in better patient outcomes.

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.

Advancing tuberculosis management: the role of predictive, preventive, and personalized medicine

Tuberculosis is a major global health issue, with approximately 10 million people falling ill and 1.4 million dying yearly. One of the most significant challenges to public health is the emergence of drug-resistant tuberculosis. For the last half-century, treating tuberculosis has adhered to a uniform management strategy in most patients. However, treatment ineffectiveness in some individuals with pulmonary tuberculosis presents a major challenge to the global tuberculosis control initiative. Unfavorable outcomes of tuberculosis treatment (including mortality, treatment failure, loss of follow-up, and unevaluated cases) may result in increased transmission of tuberculosis and the emergence of drug-resistant strains. Treatment failure may occur due to drug-resistant strains, non-adherence to medication, inadequate absorption of drugs, or low-quality healthcare. Identifying the underlying cause and adjusting the treatment accordingly to address treatment failure is important. This is where approaches such as artificial intelligence, genetic screening, and whole genome sequencing can play a critical role. In this review, we suggest a set of particular clinical applications of these approaches, which might have the potential to influence decisions regarding the clinical management of tuberculosis patients.

Lung Cancer Screening in Asia: An Expert Consensus Report

INTRODUCTION

The incidence and mortality of lung cancer are highest in Asia compared with Europe and USA, with the incidence and mortality rates being 34.4 and 28.1 per 100,000 respectively in East Asia. Diagnosing lung cancer at early stages makes the disease amenable to curative treatment and reduces mortality. In some areas in Asia, limited availability of robust diagnostic tools and treatment modalities, along with variations in specific health care investment and policies, make it necessary to have a more specific approach for screening, early detection, diagnosis, and treatment of patients with lung cancer in Asia compared with the West.

METHOD

A group of 19 advisors across different specialties from 11 Asian countries, met on a virtual Steering Committee meeting, to discuss and recommend the most affordable and accessible lung cancer screening modalities and their implementation, for the Asian population.

RESULTS

Significant risk factors identified for lung cancer in smokers in Asia include age 50 to 75 years and smoking history of more than or equal to 20 pack-years. Family history is the most common risk factor for nonsmokers. Low-dose computed tomography screening is recommended once a year for patients with screening-detected abnormality and persistent exposure to risk factors. However, for high-risk heavy smokers and nonsmokers with risk factors, reassessment scans are recommended at an initial interval of 6 to 12 months with subsequent lengthening of reassessment intervals, and it should be stopped in patients more than 80 years of age or are unable or unwilling to undergo curative treatment.

CONCLUSIONS

Asian countries face several challenges in implementing low-dose computed tomography screening, such as economic limitations, lack of efforts for early detection, and lack of specific government programs. Various strategies are suggested to overcome these challenges in Asia.

Tuberculosis prevalence after 4 years of population-wide systematic TB symptom screening and universal testing and treatment for HIV in the HPTN 071 (PopART) community-randomised trial in Zambia and South Africa: A cross-sectional survey (TREATS)

BACKGROUND

Tuberculosis (TB) prevalence remains persistently high in many settings, with new or expanded interventions required to achieve substantial reductions. The HIV Prevention Trials Network (HPTN) 071 (PopART) community-randomised trial randomised 14 communities to receive the "PopART" intervention during 2014 to 2017 (7 arm A and 7 arm B communities) and 7 communities to receive standard-of-care (arm C). The intervention was delivered door-to-door by community HIV care providers (CHiPs) and included universal HIV testing, facilitated linkage to HIV care at government health clinics, and systematic TB symptom screening. The Tuberculosis Reduction through Expanded Anti-retroviral Treatment and Screening (TREATS) study aimed to measure the impact of delivering the PopART intervention on TB outcomes, in communities with high HIV and TB prevalence.

METHODS AND FINDINGS

The study population of the HPTN 071 (PopART) trial included individuals aged ≥15 years living in 21 urban and peri-urban communities in Zambia and South Africa, with a total population of approximately 1 million and an adult HIV prevalence of around 15% at the time of the trial. Two sputum samples for TB testing were provided to CHiPs by individuals who reported ≥1 TB suggestive symptom (a cough for ≥2 weeks, unintentional weight loss ≥1.5 kg in the last month, or current night sweats) or that a household member was currently on TB treatment. Antiretroviral therapy (ART) was offered universally at clinics in arm A and according to local guidelines in arms B and C. The TREATS study was conducted in the same 21 communities as the HPTN 071 (PopART) trial between 2017 and 2022, and TB prevalence was a co-primary endpoint of the TREATS study. The primary comparison was between the PopART intervention (arms A and B combined) and the standard-of-care (arm C). During 2019 to 2021, a TB prevalence survey was conducted among randomly selected individuals aged ≥15 years (approximately 1,750 per community in arms A and B, approximately 3,500 in arm C). Participants were screened on TB symptoms and chest X-ray, with diagnostic testing using Xpert-Ultra followed by culture for individuals who screened positive. Sputum eligibility was determined by the presence of a cough for ≥2 weeks, or ≥2 of 5 "TB suggestive" symptoms (cough, weight loss for ≥4 weeks, night sweats, chest pain, and fever for ≥2 weeks), or chest X-ray CAD4TBv5 score ≥50, or no available X-ray results. TB prevalence was compared between trial arms using standard methods for cluster-randomised trials, with adjustment for age, sex, and HIV status, and multiple imputation was used for missing data on prevalent TB. Among 83,092 individuals who were eligible for the survey, 49,556 (59.6%) participated, 8,083 (16.3%) screened positive, 90.8% (7,336/8,083) provided 2 sputum samples for Xpert-Ultra testing, and 308 (4.2%) required culture confirmation. Overall, estimated TB prevalence was 0.92% (457/49,556). The geometric means of 7 community-level prevalence estimates were 0.91%, 0.70%, and 0.69% in arms A, B, and C, respectively, with no evidence of a difference comparing arms A and B combined with arm C (adjusted prevalence ratio 1.14, 95% confidence interval, CI [0.67, 1.95], p = 0.60). TB prevalence was higher among people living with HIV than HIV-negative individuals, with an age-sex-community adjusted odds ratio of 2.29 [95% CI 1.54, 3.41] in Zambian communities and 1.61 [95% CI 1.13, 2.30] in South African communities. The primary limitations are that the study was powered to detect only large reductions in TB prevalence in the intervention arm compared with standard-of-care, and the between-community variation in TB prevalence was larger than anticipated.

CONCLUSIONS

There was no evidence that the PopART intervention reduced TB prevalence. Systematic screening for TB that is based on symptom screening alone may not be sufficient to achieve a large reduction in TB prevalence over a period of several years. Including chest X-ray screening alongside TB symptom screening could substantially increase the sensitivity of systematic screening for TB.

TRIAL REGISTRATION

The TREATS study was registered with ClinicalTrials.gov Identifier: NCT03739736 on November 14, 2018. The HPTN 071 (PopART) trial was registered at ClinicalTrials.gov under number NCT01900977 on July 17, 2013.

A prospective multicenter clinical research study validating the effectiveness and safety of a chest X-ray-based pulmonary tuberculosis screening software JF CXR-1 built on a convolutional neural network algorithm

Background

Chest radiography (chest X-ray or CXR) plays an important role in the early detection of active pulmonary tuberculosis (TB). In areas with a high TB burden that require urgent screening, there is often a shortage of radiologists available to interpret the X-ray results. Computer-aided detection (CAD) software employed with artificial intelligence (AI) systems may have the potential to solve this problem.

Objective

We validated the effectiveness and safety of pulmonary tuberculosis imaging screening software that is based on a convolutional neural network algorithm.

Methods

We conducted prospective multicenter clinical research to validate the performance of pulmonary tuberculosis imaging screening software (JF CXR-1). Volunteers under the age of 15 years, both with or without suspicion of pulmonary tuberculosis, were recruited for CXR photography. The software reported a probability score of TB for each participant. The results were compared with those reported by radiologists. We measured sensitivity, specificity, consistency rate, and the area under the receiver operating characteristic curves (AUC) for the diagnosis of tuberculosis. Besides, adverse events (AE) and severe adverse events (SAE) were also evaluated.

Results

The clinical research was conducted in six general infectious disease hospitals across China. A total of 1,165 participants were enrolled, and 1,161 were enrolled in the full analysis set (FAS). Men accounted for 60.0% (697/1,161). Compared to the results from radiologists on the board, the software showed a sensitivity of 94.2% (95% CI: 92.0-95.8%) and a specificity of 91.2% (95% CI: 88.5-93.2%). The consistency rate was 92.7% (91.1-94.1%), with a Kappa value of 0.854 (P = 0.000). The AUC was 0.98. In the safety set (SS), which consisted of 1,161 participants, 0.3% (3/1,161) had AEs that were not related to the software, and no severe AEs were observed.

Conclusion

The software for tuberculosis screening based on a convolutional neural network algorithm is effective and safe. It is a potential candidate for solving tuberculosis screening problems in areas lacking radiologists with a high TB burden.

Impact of a multi-disease integrated screening and diagnostic model for COVID-19, TB, and HIV in Lesotho

The surge of the COVID-19 pandemic challenged health services globally, and in Lesotho, the HIV and tuberculosis (TB) services were similarly affected. Integrated, multi-disease diagnostic services were proposed solutions to mitigate these disruptions. We describe and evaluate the effect of an integrated, hospital-based COVID-19, TB and HIV screening and diagnostic model in two rural districts in Lesotho, during the period between December 2020 and August 2022. Adults, hospital staff, and children above 5 years attending two hospitals were pre-screened for COVID-19 and TB symptoms. After a positive pre-screening, participants were offered to enroll in a service model that included clinical evaluation, chest radiography, SARS-CoV-2, TB, and HIV testing. Participants diagnosed with COVID-19, TB, or HIV were contacted after 28 days to evaluate their health status and linkage to HIV and/or TB care services. Of the 179160 participants pre-screened, 6623(3.7%) pre-screened positive, and 4371(66%) were enrolled in this service model. Of the total 458 diagnoses, only 17 happened in children. One positive rapid antigen test for SARS-CoV-2 was found per 11 participants enrolled, one Xpert-positive TB case was diagnosed per 85 people enrolled, and 1 new HIV diagnosis was done per 182 people enrolled. Of the 321(82.9%) participants contacted after 28 days of diagnosis, 304(94.7%) reported to be healthy. Of the individuals that were newly diagnosed with HIV or TB, 18/24(75.0%) and 46/51(90.1%) started treatment within 28 days of the diagnosis. This screening and diagnostic model successfully maintained same-day, integrated COVID-19, TB, and HIV testing services, despite frequent disruptions caused by the surge of COVID-19 waves, healthcare seeking patterns, and the volatile context (social measures, travel restrictions, population lockdowns). There were positive effects in avoiding diagnostic delays and ensuring linkage to services, however, diagnostic yields for adults and children were low. To inform future preparedness plans, research will need to identify essential health interventions and how to optimize them along each phase of the emergency response.

Artificial Intelligence in Paediatric Tuberculosis

Tuberculosis (TB) continues to be a leading cause of death in children despite global efforts focused on early diagnosis and interventions to limit the spread of the disease. This challenge has been made more complex in the context of the coronavirus pandemic, which has disrupted the "End TB Strategy" and framework set out by the World Health Organization (WHO). Since the inception of artificial intelligence (AI) more than 60 years ago, the interest in AI has risen and more recently we have seen the emergence of multiple real-world applications, many of which relate to medical imaging. Nonetheless, real-world AI applications and clinical studies are limited in the niche area of paediatric imaging. This review article will focus on how AI, or more specifically deep learning, can be applied to TB diagnosis and management in children. We describe how deep learning can be utilised in chest imaging to provide computer-assisted diagnosis to augment workflow and screening efforts. We also review examples of recent AI applications for TB screening in resource constrained environments and we explore some of the challenges and the future directions of AI in paediatric TB.

Diagnostic accuracy of three computer-aided detection systems for detecting pulmonary tuberculosis on chest radiography when used for screening: Analysis of an international, multicenter migrants screening study

The aim of this study was to independently evaluate the diagnostic accuracy of three artificial intelligence (AI)-based computer aided detection (CAD) systems for detecting pulmonary tuberculosis (TB) on global migrants screening chest x-ray (CXR) cases when compared against both microbiological and radiological reference standards (MRS and RadRS, respectively). Retrospective clinical data and CXR images were collected from the International Organization for Migration (IOM) pre-migration health assessment TB screening global database for US-bound migrants. A total of 2,812 participants were included in the dataset used for analysis against RadRS, of which 1,769 (62.9%) had accompanying microbiological test results and were included against MRS. All CXRs were interpreted by three CAD systems (CAD4TB v6, Lunit INSIGHT v4.9.0, and qXR v2) in offline setting, and re-interpreted by two expert radiologists in a blinded fashion. The performance was evaluated using receiver operating characteristics curve (ROC), estimates of sensitivity and specificity at different CAD thresholds against both microbiological and radiological reference standards (MRS and RadRS, respectively), and was compared with that of the expert radiologists. The area under the curve against MRS was highest for Lunit (0.85; 95% CI 0.83-0.87), followed by qXR (0.75; 95% CI 0.72-0.77) and then CAD4TB (0.71; 95% CI 0.68-0.73). At a set specificity of 70%, Lunit had the highest sensitivity (81.4%; 95% CI 77.9-84.6); at a set sensitivity of 90%, specificity was also highest for Lunit (54.5%; 95% CI 51.7-57.3). The CAD systems performed comparable to the sensitivity (98.3%), and except CAD4TB, to specificity (13.7%) of the expert radiologists. Similar trends were observed when using RadRS. Area under the curve against RadRS was highest for CAD4TB (0.87; 95% CI 0.86-0.89) and Lunit (0.87; 95% CI 0.85-0.88) followed by qXR (0.81; 95% CI 0.80-0.83). At a set specificity of 70%, CAD4TB had highest sensitivity (84.1%; 95% CI 82.3-85.8) followed by Lunit (80.9%; 95% CI 78.9-82.7); and at a set sensitivity of 90%, specificity was also highest for CAD4TB (54.6%; 95% CI 51.3-57.8). In conclusion, the study demonstrated that the three CAD systems had broadly similar diagnostic accuracy with regard to TB screening and comparable accuracy to an expert radiologist against MRS. Compared with different reference standards, Lunit performed better than both qXR and CAD4TB against MRS, and CAD4TB and Lunit better than qXR against RadRS. Moreover, the performance of the CADs can be impacted by characteristics of subgroup of population. The main limitation was that our study relied on retrospective data and MRS was not routinely done in individuals with a low suspicion of TB and a normal CXR. Our findings suggest that CAD systems could be a useful tool for TB screening programs in remote, high TB prevalent places where access to expert radiologists may be limited. However, further large-scale prospective studies are needed to address outstanding questions around the operational performance and technical requirements of the CAD systems.

Utilization of artificial intelligence for tuberculosis screening in Nepal

BACKGROUNDS

Tuberculosis (TB) is an infectious disease that needs to be diagnosed and enrolled for treatment. Artificial intelligence for TB (AI4TB) software screens TB suspected cases at the point of care and helps in quick diagnosis. This study aims to explore the significance and usefulness of AI4TB by comparing its performance with different diagnostic test results.

METHODS

A cross-sectional study was conducted among 197 participants who had symptoms suggestive to TB. The chest X-ray images were analyzed by AI4TB software and human expert readers. The bacteriological test results were obtained, and Kappa test was applied to calculate the inter-reader reliability. The sensitivity, specificity, positive predictive value and negative predictive value was calculated and ROC curve was generated.

RESULTS

Among 85 sputum smear microscopy, about 21% of the had sputum positivity rate. At 0.4 threshold: 62.4%, at 0.5 threshold: 58.4% and at 0.6 threshold: 50.3% symptoms suggestive cases were identified having abnormal X-ray images. Reader-I identified 28.4% and Reader-II identified 37.1% of the symptoms suggestive cases of TB as positive cases. There was a significant substantial agreement between two human expert readers (k-0.783, p-value: <0.001). The ROC curve explored the fair sensitivity accuracy of the AI4TB test results at 0.5 threshold level (AUC = 0.72) and at 0.6 threshold level (AUC = 0.77).

CONCLUSION

The sensitivity of the AI4TB was higher compared to different human readers. AI4TB can be the relevant screening tool for the TB symptoms suggestive cases prior to the laboratory test in the countries like Nepal with deficient health manpower.

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.

Artificial Intelligence-Based Software with CE Mark for Chest X-ray Interpretation: Opportunities and Challenges

Chest X-ray (CXR) is the most important technique for performing chest imaging, despite its well-known limitations in terms of scope and sensitivity. These intrinsic limitations of CXR have prompted the development of several artificial intelligence (AI)-based software packages dedicated to CXR interpretation. The online database "AI for radiology" was queried to identify CE-marked AI-based software available for CXR interpretation. The returned studies were divided according to the targeted disease. AI-powered computer-aided detection software is already widely adopted in screening and triage for pulmonary tuberculosis, especially in countries with few resources and suffering from high a burden of this disease. AI-based software has also been demonstrated to be valuable for the detection of lung nodules detection, automated flagging of positive cases, and post-processing through the development of digital bone suppression software able to produce digital bone suppressed images. Finally, the majority of available CE-marked software packages for CXR are designed to recognize several findings, with potential differences in sensitivity and specificity for each of the recognized findings.

Trends and challenges of multi-drug resistance in childhood tuberculosis

Drug-resistant tuberculosis (DR-TB) in children is a growing global health concern, This review provides an overview of the current epidemiology of childhood TB and DR-TB, including prevalence, incidence, and mortality. We discuss the challenges in diagnosing TB and DR-TB in children and the limitations of current diagnostic tools. We summarize the challenges associated with treating multi-drug resistance TB in childhood, including limitations of current treatment options, drug adverse effects, prolonged regimens, and managing and monitoring during treatment. We highlight the urgent need for improved diagnosis and treatment of DR-TB in children. The treatment of children with multidrug-resistant tuberculosis will be expanded to include the evaluation of new drugs or new combinations of drugs. Basic research is needed to support the technological development of biomarkers to assess the phase of therapy, as well as the urgent need for improved diagnostic and treatment options.

Optimising computer aided detection to identify intra-thoracic tuberculosis on chest x-ray in South African children

Diagnostic tools for paediatric tuberculosis remain limited, with heavy reliance on clinical algorithms which include chest x-ray. Computer aided detection (CAD) for tuberculosis on chest x-ray has shown promise in adults. We aimed to measure and optimise the performance of an adult CAD system, CAD4TB, to identify tuberculosis on chest x-rays from children with presumptive tuberculosis. Chest x-rays from 620 children <13 years enrolled in a prospective observational diagnostic study in South Africa, were evaluated. All chest x-rays were read by a panel of expert readers who attributed each with a radiological reference of either 'tuberculosis' or 'not tuberculosis'. Of the 525 chest x-rays included in this analysis, 80 (40 with a reference of 'tuberculosis' and 40 with 'not tuberculosis') were allocated to an independent test set. The remainder made up the training set. The performance of CAD4TB to identify 'tuberculosis' versus 'not tuberculosis' on chest x-ray against the radiological reference read was calculated. The CAD4TB software was then fine-tuned using the paediatric training set. We compared the performance of the fine-tuned model to the original model. Our findings were that the area under the receiver operating characteristic curve (AUC) of the original CAD4TB model, prior to fine-tuning, was 0.58. After fine-tuning there was an improvement in the AUC to 0.72 (p = 0.0016). In this first-ever description of the use of CAD to identify tuberculosis on chest x-ray in children, we demonstrate a significant improvement in the performance of CAD4TB after fine-tuning with a set of well-characterised paediatric chest x-rays. CAD has the potential to be a useful additional diagnostic tool for paediatric tuberculosis. We recommend replicating the methods we describe using a larger chest x-ray dataset from a more diverse population and evaluating the potential role of CAD to replace a human-read chest x-ray within treatment-decision algorithms for paediatric tuberculosis.

Artificial intelligence-based computer aided detection (AI-CAD) in the fight against tuberculosis: Effects of moving health technologies in global health

Computer Aided Detection software based on Artificial Intelligence (AI-CAD), combined with chest X-rays have recently been promoted as an easy fix for a complex problem: ending TB by 2030. WHO has recommended the use of such imaging devices in 2021 and many partnerships have helped propose benchmark analysis and technology comparisons to facilitate their "market access". Our aim is to examine the socio-political and health issues that stem from using AI-CAD technology in a global health context conceptualized as a set of practice and ideas organizing global intervention "in the life of others". We also question how this technology, which is not yet fully implemented in routine use, may limit or amplify some inequalities in the care of tuberculosis. We describe AI-CAD through Actor-Network-Theory framework to understand the global assemblage and composite activities associated with detection through AI-CAD, and interrogate how the technology itself may consolidate a specific configuration of "global health". We explore the various dimensions of AI-CAD "health effects model": technology design, development, regulation, institutional competition, social interaction and health cultures. On a broader level, AI-CAD represents a new version of global health's accelerationist model centered on "moving and autonomous-presumed technologies". We finally present key aspects in our research which help discuss the theories mobilized: AI-CAD ambivalent insertion in global health, the social lives of its data: from efficacy to markets and AI-CAD human care and maintenance it requires. We reflect on the conditions that will affect AI-CAD use and its promises. In the end, the risk of new detection technologies such as AI-CAD is indeed that the fight against TB could be reduced to one that is purely technical and technological, with neglect to its social determinants and effects.

Identification of Active Pulmonary Tuberculosis Among Patients With Positive Interferon-Gamma Release Assay Results: Value of a Deep Learning-based Computer-aided Detection System in Different Scenarios of Implementation

PURPOSE

To evaluate the accuracy of a deep learning-based computer-aided detection (CAD) system in identifying active pulmonary tuberculosis on chest radiographs (CRs) of patients with positive interferon-gamma release assay (IGRA) results in different scenarios of clinical implementation.

MATERIALS AND METHODS

We collected the CRs of consecutive patients with positive IGRA results. Findings of active pulmonary tuberculosis on CRs were independently evaluated by the CAD and a thoracic radiologist, followed by interpretation using the CAD. Sensitivity and specificity were evaluated in different scenarios: (a) radiologists' interpretation, (b) radiologists' CAD-assisted interpretation, and (c) CAD-based prescreening (radiologists' interpretation for positive CAD results only). We conducted a reader test to compare the accuracy of the CAD with those of 5 radiologists.

RESULTS

Among 1780 patients (men, 53.8%; median age, 56 y), 44 (2.5%) were diagnosed with active pulmonary tuberculosis. The CAD-assisted interpretation exhibited a higher sensitivity (81.8% vs. 72.7%; P =0.046) but lower specificity than the radiologists' interpretation (84.1% vs. 85.7%; P <0.001). The CAD-based prescreening exhibited a higher specificity than the radiologists' interpretation (88.8% vs. 85.7%; P <0.001) at the same sensitivity, with a workload reduction of 85.2% (1780 to 263). In the reader test, the CAD exhibited a higher sensitivity than radiologists (72.7% vs. 59.5%; P =0.005) at the same specificity (88.0%), and CAD-assisted interpretation significantly improved the sensitivity of radiologists' interpretation (72.3%; P <0.001).

CONCLUSIONS

For identifying active pulmonary tuberculosis among patients with positive IGRA results, deep learning-based CAD can enhance the sensitivity of interpretation. CAD-based prescreening may reduce the radiologists' workload at an improved specificity.

Updating the WHO target product profile for next-generation Mycobacterium tuberculosis drug susceptibility testing at peripheral centres

There were approximately 10 million tuberculosis (TB) cases in 2020, of which 500,000 were drug-resistant. Only one third of drug-resistant TB cases were diagnosed and enrolled on appropriate treatment, an issue partly driven by a lack of rapid, accurate drug-susceptibility testing (DST) tools deployable in peripheral settings. In 2014, World Health Organization (WHO) published target product profiles (TPPs) which detailed minimal and optimal criteria to address high-priority TB diagnostic needs, including DST. Since then, the TB community's needs have evolved; new treatment regimens, changes in TB definitions, further emergence of drug resistance, technological advances, and changing end-users requirements have necessitated an update. The DST TPP's revision was therefore undertaken by WHO with the Stop TB Partnership New Diagnostics Working Group. We describe the process of updating the TPP for next-generation TB DST for use at peripheral centres, highlight key updates, and discuss guidance regarding technical and operational specifications.

Clinical assistant decision-making model of tuberculosis based on electronic health records

BACKGROUND

Tuberculosis is a dangerous infectious disease with the largest number of reported cases in China every year. Preventing missed diagnosis has an important impact on the prevention, treatment, and recovery of tuberculosis. The earliest pulmonary tuberculosis prediction models mainly used traditional image data combined with neural network models. However, a single data source tends to miss important information, such as primary symptoms and laboratory test results, that is available in multi-source data like medical records and tests. In this study, we propose a multi-stream integrated pulmonary tuberculosis diagnosis model based on structured and unstructured multi-source data from electronic health records. With the limited number of lung specialists and the high prevalence of tuberculosis, the application of this auxiliary diagnosis model can make substantial contributions to clinical settings.

METHODS

The subjects were patients at the respiratory department and infectious cases department of a large comprehensive hospital in China between 2015 to 2020. A total of 95,294 medical records were selected through a quality control process. Each record contains structured and unstructured data. First, numerical expressions of features for structured data were created. Then, feature engineering was performed through decision tree model, random forest, and GBDT. Features were included in the feature exclusion set as per their weights in descending order. When the importance of the set was higher than 0.7, this process was concluded. Finally, the contained features were used for model training. In addition, the unstructured free-text data was segmented at the character level and input into the model after indexing. Tuberculosis prediction was conducted through a multi-stream integration tuberculosis diagnosis model (MSI-PTDM), and the evaluation indices of accuracy, AUC, sensitivity, and specificity were compared against the prediction results of XGBoost, Text-CNN, Random Forest, SVM, and so on.

RESULTS

Through a variety of characteristic engineering methods, 20 characteristic factors, such as main complaint hemoptysis, cough, and test erythrocyte sedimentation rate, were selected, and the influencing factors were analyzed using the Chinese diagnostic standard of pulmonary tuberculosis. The area under the curve values for MSI-PTDM, XGBoost, Text-CNN, RF, and SVM were 0.9858, 0.9571, 0.9486, 0.9428, and 0.9429, respectively. The sensitivity, specificity, and accuracy of MSI-PTDM were 93.18%, 96.96%, and 96.96%, respectively. The MSI-PTDM prediction model was installed at a doctor workstation and operated in a real clinic environment for 4 months. A total of 692,949 patients were monitored, including 484 patients with confirmed pulmonary tuberculosis. The model predicted 440 cases of pulmonary tuberculosis. The positive sample recognition rate was 90.91%, the false-positive rate was 9.09%, the negative sample recognition rate was 96.17%, and the false-negative rate was 3.83%.

CONCLUSIONS

MSI-PTDM can process sparse data, dense data, and unstructured text data concurrently. The model adds a feature domain vector embedding the medical sparse features, and the single-valued sparse vectors are represented by multi-dimensional dense hidden vectors, which not only enhances the feature expression but also alleviates the side effects of sparsity on the model training. However, there may be information loss when features are extracted from text, and adding the processing of original unstructured text makes up for the error within the above process to a certain extent, so that the model can learn data more comprehensively and effectively. In addition, MSI-PTDM also allows interaction between features, considers the combination effect between patient features, adds more complex nonlinear calculation considerations, and improves the learning ability of the model. It has been verified using a test set and via deployment within an actual outpatient environment.

Evaluation of tuberculosis diagnostic test accuracy using Bayesian latent class analysis in the presence of conditional dependence between the diagnostic tests used in a community-based tuberculosis screening study

Diagnostic accuracy studies in pulmonary tuberculosis (PTB) are complicated by the lack of a perfect reference standard. This limitation can be handled using latent class analysis (LCA), assuming independence between diagnostic test results conditional on the true unobserved PTB status. Test results could remain dependent, however, e.g. with diagnostic tests based on a similar biological basis. If ignored, this gives misleading inferences. Our secondary analysis of data collected during the first year (May 2018 -May 2019) of a community-based multi-morbidity screening program conducted in the rural uMkhanyakude district of KwaZulu Natal, South Africa, used Bayesian LCA. Residents of the catchment area, aged ≥15 years and eligible for microbiological testing, were analyzed. Probit regression methods for dependent binary data sequentially regressed each binary test outcome on other observed test results, measured covariates and the true unobserved PTB status. Unknown model parameters were assigned Gaussian priors to evaluate overall PTB prevalence and diagnostic accuracy of 6 tests used to screen for PTB: any TB symptom, radiologist conclusion, Computer Aided Detection for TB version 5 (CAD4TBv5≥53), CAD4TBv6≥53, Xpert Ultra (excluding trace) and culture. Before the application of our proposed model, we evaluated its performance using a previously published childhood pulmonary TB (CPTB) dataset. Standard LCA assuming conditional independence yielded an unrealistic prevalence estimate of 18.6% which was not resolved by accounting for conditional dependence among the true PTB cases only. Allowing, also, for conditional dependence among the true non-PTB cases produced a 1.1% plausible prevalence. After incorporating age, sex, and HIV status in the analysis, we obtained 0.9% (95% CrI: 0.6, 1.3) overall prevalence. Males had higher PTB prevalence compared to females (1.2% vs. 0.8%). Similarly, HIV+ had a higher PTB prevalence compared to HIV- (1.3% vs. 0.8%). The overall sensitivity for Xpert Ultra (excluding trace) and culture were 62.2% (95% CrI: 48.7, 74.4) and 75.9% (95% CrI: 61.9, 89.2), respectively. Any chest X-ray abnormality, CAD4TBv5≥53 and CAD4TBv6≥53 had similar overall sensitivity. Up to 73.3% (95% CrI: 61.4, 83.4) of all true PTB cases did not report TB symptoms. Our flexible modelling approach yields plausible, easy-to-interpret estimates of sensitivity, specificity and PTB prevalence under more realistic assumptions. Failure to fully account for diagnostic test dependence can yield misleading inferences.