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

Qin, Zhi Zhen; Naheyan, Tasneem; Rühwald, Morten; Denkinger, Claudia M.; Gelaw, Sifrash Meseret; Nash, Madlen; Creswell, Jacob; Kik, Sandra V.

doi:10.1016/j.tube.2020.102049

Cited by 36 publications

(37 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…CAD4TB is a proprietary algorithm and the demographic and clinical characteristics of the training data are unpublished. 37 Therefore it is unclear to what extent community-diagnosed asymptomatic TB has been used to train CAD4TB. It would be interesting to see if CAD4TB and other CAD-algorithms can be refined to detect early, radiologically-subtle patterns of TB that is not yet clinically apparent.…”

Section: Discussionmentioning

confidence: 99%

Computer-aided interpretation of chest radiography reveals the spectrum of tuberculosis in rural South Africa

Fehr

Konigorski

Olivier

et al. 2020

Preprint

View full text Add to dashboard Cite

Background Computer-aided digital chest radiograph (CXR) interpretation can facilitate high-throughput screening for tuberculosis (TB), but its use in population-based screening has been limited. We applied an automated image interpretation algorithm, CAD4TBv5, prospectively in an HIV-endemic area. Methods Participants underwent CXR and, for those with symptoms or lung field abnormality, microbiological assessment of sputum collected at a mobile camp in rural South Africa. CAD4TBv5 scored each CXR on a 0-100 scale in the field. An expert radiologist, blinded to the CAD4TBv5 score and other data, assessed CXRs for 1) lung field abnormality and 2) findings diagnostic of active TB (R+). We estimated the performance of CAD4TBv5 for triaging (identifying lung field abnormality as a criteria for sputum examination) and diagnosis (detection of active TB as defined by microbiologic (M+) or radiologic (R+) gold standards). Findings For triaging, a CAD4TBv5 threshold of 25 identified abnormal lung fields with a sensitivity of 90.3% and specificity of 48.2%. For diagnosis, CAD4TBv5 had less agreement with the microbiological reference standard (M+) used to define definite TB (AUC 0.78) than with the radiological reference standard (R+) used to define probable TB (AUC 0.96). HIV-serostatus did not impact CAD4TB's performance. Interpretation A low CAD4TBv5 threshold was required to achieve acceptable triaging sensitivity. Low specificity at this threshold led to high rates of sputum collection despite normal lung fields. CAD4TBv5 had difficulty identifying microbiologically-confirmed TB cases with subtle radiological features but had excellent agreement with the radiologist in identifying radiologically-defined TB cases. We conclude that computer-aided CXR interpretation can be useful in population-based screening in HIV-endemic settings, but threshold selection should be guided by setting-specific piloting and priorities. CXR interpretation algorithms require refinement for the identification of radiologically-subtle early TB.

show abstract

Section: Discussionmentioning

confidence: 99%

Computer-aided interpretation of chest radiography reveals the spectrum of tuberculosis in rural South Africa

Fehr

Konigorski

Olivier

et al. 2020

Preprint

View full text Add to dashboard Cite

show abstract

“…8 Several commercial AI algorithms have emerged in recent years promising to identify tuberculosis-related abnormalities from digital chest x-ray images. 9 AI algorithms produce a continuous abnormality score (from 0 to 100 or from 0 to 1) that represents the likelihood of the presence of tuberculosis-associated abnormalities. 10 Although some software comes with preset threshold abnormality scores, all algorithms also allow users to customise the threshold score at any level to dichotomise the output into binary classifications (either suggesting confirmatory testing for tuberculosis or not).…”

Section: Introductionmentioning

confidence: 99%

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

Qin¹,

Ahmed

Sarker

et al. 2021

The Lancet Digital Health

Self Cite

127

123

View full text Add to dashboard Cite

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

show abstract

“…[9,10]. Artificial intelligence (AI)-assisted interpretation of CXRs recently obtained WHO policy recommendation [11] and may become widespread in the coming years, reducing the requirement for skilled staff, but CXR hardware cost remains a huge barrier for access [12][13][14].…”

Section: Introductionmentioning

confidence: 99%

Diagnostic accuracy of point-of-care ultrasound for pulmonary tuberculosis: A systematic review

et al. 2021

Self Cite

View full text Add to dashboard Cite

The advent of affordable, portable ultrasound devices has led to increasing interest in the use of point-of-care ultrasound (POCUS) for the detection of pulmonary TB (PTB). We undertook a systematic review of the diagnostic accuracy of POCUS for PTB. Five databases were searched for articles published between January 2010 and June 2020. Risk of bias was assessed using QUADAS-2. Data on sensitivity and specificity of individual lung ultrasound findings were collected, with variable reference standards including PCR and sputum smear microscopy. Six of 3,919 reviewed articles were included: five in adults and one in children, with a total sample size of 564. Studies had high risk of bias in many domains. In adults, subpleural nodule and lung consolidation were the lung ultrasound findings with the highest sensitivities, ranging from 72.5% to 100.0% and 46.7% to 80.4%, respectively. Only one study reported specificity data. Variability in sensitivity may be due to variable reference standards or may imply operator dependence. There is insufficient evidence to judge the diagnostic accuracy of POCUS for PTB. There is also no consensus on the optimal protocols for acquiring and analysing POCUS images for PTB. New studies which minimise potential sources of bias are required to further assess the diagnostic accuracy of POCUS for PTB.

show abstract

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

Cited by 36 publications

References 9 publications

Computer-aided interpretation of chest radiography reveals the spectrum of tuberculosis in rural South Africa

Computer-aided interpretation of chest radiography reveals the spectrum of tuberculosis in rural South Africa

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

Diagnostic accuracy of point-of-care ultrasound for pulmonary tuberculosis: A systematic review

Contact Info

Product

Resources

About