Screening for pulmonary tuberculosis in a Tanzanian prison and computer-aided interpretation of chest X-rays

Steiner, Andreas; Mangu, Chacha; Hombergh, Jan van den; Deutekom, H. van; Ginneken, Bram van; Clowes, Petra; Mhimbira, Francis; Mfinanga, S.; Rachow, Andrea; Reither, Klaus; Höelscher, Michael

doi:10.5588/pha.15.0037

Cited by 16 publications

(15 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Similar studies, reporting diagnostic accuracy using Xpert MTB/RIF as the reference standard have been reported from Zambia in 2013 and Bangladesh in 2017 13 , 14 . Other studies from Zambia, Tanzania, South Africa and England have evaluated CAD4TB against the reference standard of culture 15 , 24 – 26 . Our current study is another data point in the series of studies, carried out in Pakistan.…”

Section: Introductionmentioning

confidence: 99%

Evaluation of the diagnostic accuracy of Computer-Aided Detection of tuberculosis on Chest radiography among private sector patients in Pakistan

Zaidi

Habib

Ginneken

et al. 2018

Sci Rep

View full text Add to dashboard Cite

The introduction of digital CXR with automated computer-aided interpretation, has given impetus to the role of CXR in TB screening, particularly in low resource, high-burden settings. The aim of this study was to evaluate the diagnostic accuracy of CAD4TB as a screening tool, implemented in the private sector in Karachi, Pakistan. This study analyzed retrospective data from CAD4TB and Xpert MTB/RIF testing carried out at two private TB treatment and diagnostic centers in Karachi. Sensitivity, specificity, potential Xperts saved, were computed and the receiver operator characteristic curves were constructed for four different models of CAD4TB. A total of 6,845 individuals with presumptive TB were enrolled in the study, 15.2% of which had MTB + ve result on Xpert. A high sensitivity (range 65.8–97.3%) and NPV (range 93.1–98.4%) were recorded for CAD4TB. The Area under the ROC curve (AUC) for CAD4TB was 0.79. CAD4TB with patient demographics (age and gender) gave an AUC of 0.83. CAD4TB offered high diagnostic accuracy. In low resource settings, CAD4TB, as a triage tool could minimize use of Xpert. Using CAD4TB in combination with age and gender data enhanced the performance of the software. Variations in demographic information generate different individual risk probabilities for the same CAD4TB scores.

show abstract

Section: Introductionmentioning

confidence: 99%

Evaluation of the diagnostic accuracy of Computer-Aided Detection of tuberculosis on Chest radiography among private sector patients in Pakistan

Zaidi

Habib

Ginneken

et al. 2018

Sci Rep

View full text Add to dashboard Cite

show abstract

“…Studies have observed better correlation between culture and radiologist findings. 10,11 Thus, even without culture results, our findings raise several programmatic implications. First, almost a quarter of patients may be over-treated during Tas4TB.…”

Section: Discussionmentioning

confidence: 79%

“…15 By contrast, there was a moderate agreement in ratings between Radiologist A and Radiologist B, a finding that is consistent with most studies that suggest radiologist ratings to be more reliable than those by medical officers and CAD4TB (Table 5). 11,12 The strengths of this study lie in the fact that we purposively selected expert radiologists, based on their experience in rating CXRs; also, study radiologists were blinded to the scores of the other radiologists and independent ratings to ensure credibility and accuracy of results. We facilitated their ratings by providing information on HIV status on the rating proforma.…”

Section: Discussionmentioning

confidence: 99%

“…However, where ratings were done by radiologists, studies have reported higher sensitivities and specificities than both medical officers and computer-aided diagnosis for TB (CAD4TB) (Delft Imaging Systems, Neenendaal, The Netherlands). [10][11][12] The reliability of field CXR scores is not known in Zimbabwe, as medical officers do not routinely undergo assessments for CXR competence. Accurate CXR interpretation is crucial as it acts as a 'gatekeeper', filtering patients who then proceed to Xpert testing or TB treatment.…”

Section: Public Health Actionmentioning

confidence: 99%

See 1 more Smart Citation

Targeted active screening for tuberculosis in Zimbabwe: are field digital chest X-ray ratings reliable?

Timire

Sandy

Ngwenya

et al. 2019

public health action

View full text Add to dashboard Cite

Setting: Fifteen purposively selected districts in Zimbabwe in which targeted active screening for tuberculosis (Tas4TB) was conducted among TB high-risk groups (HRGs). There were 230 patients started on TB treatment on the basis of chest X-ray (CXR) results without corresponding bacteriological confirmation.Objectives: To determine 1) the percentage of agreements in digital CXR ratings by medical officers against final ratings by radiologist(s), 2) inter-rater agreement in CXR ratings between medical officers and radiologists, and 3) number (and proportion) of patients belonging to HRGs who were over-treated during Tas4TB.Design: This was a cross-sectional study using programme data.Results: A total of 168 patients had their CXRs rated by two independent radiologists. Discordances among the radiologists were resolved by a third index radiologist, who provided the final rating. κ scores were 0.01 (field ratings vs. Radiologist A); 0.02 (field ratings vs. Radiologist B); 0.74 (Radiologists A vs. B). The percentage agreement for field and final radiologist rating was 70% (95%CI 64–78). Around 29% (95%CI 23–36) of the patients were potentially over-treated during Tas4TB.Conclusion: Over a quarter of patients with presumptive TB are potentially over-treated during Tas4TB. Over-treatment is highest among those with previous contact with TB patients. Trainings of radiographers and medical officers may improve CXR ratings.

show abstract

“…The Computer-Aided Detection for Tuberculosis (CAD4TB), SemanticMD, and Qure.ai are selected examples of currently available computer-aided detection (CAD) software. CAD4TB is TB specific and had demonstrated a good diagnostic performance [ 6 , 7 ] but still inferior to that of expert readers [ 8 ]. As of 2016, WHO provided no recommendations on using CAD for TB because of the small number of studies, methodological limitations, and limited generalizability of the findings [ 1 ].…”

Section: Introductionmentioning

confidence: 99%

Deep learning for automated classification of tuberculosis-related chest X-Ray: dataset distribution shift limits diagnostic performance generalizability

Sathitratanacheewin

Sunanta

Pongpirul

2020

Heliyon

View full text Add to dashboard Cite

Background Machine learning has been an emerging tool for various aspects of infectious diseases including tuberculosis surveillance and detection. However, the World Health Organization (WHO) provided no recommendations on using computer-aided tuberculosis detection software because of a small number of studies, methodological limitations, and limited generalizability of the findings. Methods To quantify the generalizability of the machine-learning model, we developed a Deep Convolutional Neural Network (DCNN) model using a Tuberculosis (TB)-specific chest x-ray (CXR) dataset of one population (National Library of Medicine Shenzhen No.3 Hospital) and tested it with non-TB-specific CXR dataset of another population (National Institute of Health Clinical Centers). Results In the training and intramural test sets using the Shenzhen hospital database, the DCCN model exhibited an AUC of 0.9845 and 0.8502 for detecting TB, respectively. However, the AUC of the supervised DCNN model in the ChestX-ray8 dataset was dramatically dropped to 0.7054. Using the cut points at 0.90, which suggested 72% sensitivity and 82% specificity in the Shenzhen dataset, the final DCNN model estimated that 36.51% of abnormal radiographs in the ChestX-ray8 dataset were related to TB. Conclusion A supervised deep learning model developed by using the training dataset from one population may not have the same diagnostic performance in another population. Conclusion: Technical specification of CXR images, disease severity distribution, dataset distribution shift, and overdiagnosis should be examined before implementation in other settings.

show abstract

Screening for pulmonary tuberculosis in a Tanzanian prison and computer-aided interpretation of chest X-rays

Cited by 16 publications

References 14 publications

Evaluation of the diagnostic accuracy of Computer-Aided Detection of tuberculosis on Chest radiography among private sector patients in Pakistan

Evaluation of the diagnostic accuracy of Computer-Aided Detection of tuberculosis on Chest radiography among private sector patients in Pakistan

Targeted active screening for tuberculosis in Zimbabwe: are field digital chest X-ray ratings reliable?

Deep learning for automated classification of tuberculosis-related chest X-Ray: dataset distribution shift limits diagnostic performance generalizability

Contact Info

Product

Resources

About