Fusion of local and global detection systems to detect tuberculosis in chest radiographs

Chest X-ray Bone Suppression for Improving Classification of Tuberculosis-Consistent Findings

Chest X-rays (CXRs) are the most commonly performed diagnostic examination to detect cardiopulmonary abnormalities. However, the presence of bony structures such as ribs and clavicles can obscure subtle abnormalities, resulting in diagnostic errors. This study aims to build a deep learning (DL)-based bone suppression model that identifies and removes these occluding bony structures in frontal CXRs to assist in reducing errors in radiological interpretation, including DL workflows, related to detecting manifestations consistent with tuberculosis (TB). Several bone suppression models with various deep architectures are trained and optimized using the proposed combined loss function and their performances are evaluated in a cross-institutional test setting using several metrics such as mean absolute error (MAE), peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and multiscale structural similarity measure (MS-SSIM). The best-performing model (ResNet-BS) (PSNR = 34.0678; MS-SSIM = 0.9828) is used to suppress bones in the publicly available Shenzhen and Montgomery TB CXR collections. A VGG-16 model is pretrained on a large collection of publicly available CXRs. The CXR-pretrained model is then fine-tuned individually on the non-bone-suppressed and bone-suppressed CXRs of Shenzhen and Montgomery TB CXR collections to classify them as showing normal lungs or TB manifestations. The performances of these models are compared using several performance metrics such as accuracy, the area under the curve (AUC), sensitivity, specificity, precision, F-score, and Matthews correlation coefficient (MCC), analyzed for statistical significance, and their predictions are qualitatively interpreted through class-selective relevance maps (CRMs). It is observed that the models trained on bone-suppressed CXRs (Shenzhen: AUC = 0.9535 ± 0.0186; Montgomery: AUC = 0.9635 ± 0.0106) significantly outperformed (p < 0.05) the models trained on the non-bone-suppressed CXRs (Shenzhen: AUC = 0.8991 ± 0.0268; Montgomery: AUC = 0.8567 ± 0.0870).. Models trained on bone-suppressed CXRs improved detection of TB-consistent findings and resulted in compact clustering of the data points in the feature space signifying that bone suppression improved the model sensitivity toward TB classification.

Development and Validation of a Deep Learning-based Automatic Detection Algorithm for Active Pulmonary Tuberculosis on Chest Radiographs

Background

Detection of active pulmonary tuberculosis on chest radiographs (CRs) is critical for the diagnosis and screening of tuberculosis. An automated system may help streamline the tuberculosis screening process and improve diagnostic performance.

Methods

We developed a deep learning–based automatic detection (DLAD) algorithm using 54c221 normal CRs and 6768 CRs with active pulmonary tuberculosis that were labeled and annotated by 13 board-certified radiologists. The performance of DLAD was validated using 6 external multicenter, multinational datasets. To compare the performances of DLAD with physicians, an observer performance test was conducted by 15 physicians including nonradiology physicians, board-certified radiologists, and thoracic radiologists. Image-wise classification and lesion-wise localization performances were measured using area under the receiver operating characteristic (ROC) curves and area under the alternative free-response ROC curves, respectively. Sensitivities and specificities of DLAD were calculated using 2 cutoffs (high sensitivity [98%] and high specificity [98%]) obtained through in-house validation.

Results

DLAD demonstrated classification performance of 0.977–1.000 and localization performance of 0.973–1.000. Sensitivities and specificities for classification were 94.3%–100% and 91.1%–100% using the high-sensitivity cutoff and 84.1%–99.0% and 99.1%–100% using the high-specificity cutoff. DLAD showed significantly higher performance in both classification (0.993 vs 0.746–0.971) and localization (0.993 vs 0.664–0.925) compared to all groups of physicians.

Conclusions

Our DLAD demonstrated excellent and consistent performance in the detection of active pulmonary tuberculosis on CR, outperforming physicians, including thoracic radiologists.

Deep learning-based automated detection algorithm for active pulmonary tuberculosis on chest radiographs: diagnostic performance in systematic screening of asymptomatic individuals

OBJECTIVES

Performance of deep learning-based automated detection (DLAD) algorithms in systematic screening for active pulmonary tuberculosis is unknown. We aimed to validate DLAD algorithm for detection of active pulmonary tuberculosis and any radiologically identifiable relevant abnormality on chest radiographs (CRs) in this setting.

METHODS

We performed out-of-sample testing of a pre-trained DLAD algorithm, using CRs from 19.686 asymptomatic individuals (ages, 21.3 ± 1.9 years) as part of systematic screening for tuberculosis between January 2013 and July 2018. Area under the receiver operating characteristic curves (AUC) for diagnosis of tuberculosis and any relevant abnormalities were measured. Accuracy measures including sensitivities, specificities, positive predictive values (PPVs), and negative predictive values (NPVs) were calculated at pre-defined operating thresholds (high sensitivity threshold, 0.16; high specificity threshold, 0.46).

RESULTS

All five CRs from four individuals with active pulmonary tuberculosis were correctly classified as having abnormal findings by DLAD with specificities of 0.959 and 0.997, PPVs of 0.006 and 0.068, and NPVs of both 1.000 at high sensitivity and high specificity thresholds, respectively. With high specificity thresholds, DLAD showed comparable diagnostic measures with the pooled radiologists (p values > 0.05). For the radiologically identifiable relevant abnormality (n = 28), DLAD showed an AUC value of 0.967 (95% confidence interval, 0.938-0.996) with sensitivities of 0.821 and 0.679, specificities of 0.960 and 0.997, PPVs of 0.028 and 0.257, and NPVs of both 0.999 at high sensitivity and high specificity thresholds, respectively.

CONCLUSIONS

In systematic screening for tuberculosis in a low-prevalence setting, DLAD algorithm demonstrated excellent diagnostic performance, comparable with the radiologists in the detection of active pulmonary tuberculosis.

KEY POINTS

• Deep learning-based automated detection algorithm detected all chest radiographs with active pulmonary tuberculosis with high specificities and negative predictive values in systematic screening. • Deep learning-based automated detection algorithm had comparable diagnostic measures with the radiologists for detection of active pulmonary tuberculosis on chest radiographs. • For the detection of radiologically identifiable relevant abnormalities on chest radiographs, deep learning-based automated detection algorithm showed excellent diagnostic performance in systematic screening.

Computer-Aided System for the Detection of Multicategory Pulmonary Tuberculosis in Radiographs

The early screening and diagnosis of tuberculosis plays an important role in the control and treatment of tuberculosis infections. In this paper, an integrated computer-aided system based on deep learning is proposed for the detection of multiple categories of tuberculosis lesions in chest radiographs. In this system, the fully convolutional neural network method is used to segment the lung area from the entire chest radiograph for pulmonary tuberculosis detection. Different from the previous analysis of the whole chest radiograph, we focus on the specific tuberculosis lesion areas for the analysis and propose the first multicategory tuberculosis lesion detection method. In it, a learning scalable pyramid structure is introduced into the Faster Region-based Convolutional Network (Faster RCNN), which effectively improves the detection of small-area lesions, mines indistinguishable samples during the training process, and uses reinforcement learning to reduce the detection of false-positive lesions. To compare our method with the current tuberculosis detection system, we propose a classification rule for whole chest X-rays using a multicategory tuberculosis lesion detection model and achieve good performance on two public datasets (Montgomery: AUC = 0.977 and accuracy = 0.926; Shenzhen: AUC = 0.941 and accuracy = 0.902). Our proposed computer-aided system is superior to current systems that can be used to assist radiologists in diagnoses and public health providers in screening for tuberculosis in areas where tuberculosis is endemic.

Deep Learning-Based Decision-Tree Classifier for COVID-19 Diagnosis From Chest X-ray Imaging

The global pandemic of coronavirus disease 2019 (COVID-19) has resulted in an increased demand for testing, diagnosis, and treatment. Reverse transcription polymerase chain reaction (RT-PCR) is the definitive test for the diagnosis of COVID-19; however, chest X-ray radiography (CXR) is a fast, effective, and affordable test that identifies the possible COVID-19-related pneumonia. This study investigates the feasibility of using a deep learning-based decision-tree classifier for detecting COVID-19 from CXR images. The proposed classifier comprises three binary decision trees, each trained by a deep learning model with convolution neural network based on the PyTorch frame. The first decision tree classifies the CXR images as normal or abnormal. The second tree identifies the abnormal images that contain signs of tuberculosis, whereas the third does the same for COVID-19. The accuracies of the first and second decision trees are 98 and 80%, respectively, whereas the average accuracy of the third decision tree is 95%. The proposed deep learning-based decision-tree classifier may be used in pre-screening patients to conduct triage and fast-track decision making before RT-PCR results are available.

Diagnosing Tuberculosis Using Deep Convolutional Neural Network

One of the global topmost causes of death is Tuberculosis (TB) which is caused by mycobacterium bacillus. The increase rate of infected people and the recorded deaths from TB disease is as a result of its transmissibility, lack of early diagnosis, and inadequate professional radiologist in developing regions where TB is more prevalent. Tuberculosis is unquestionably curable but needs to be detected early for necessary treatment to be effective. Many screening techniques are available, but chest radiograph has proven to be valuable for screening pulmonary diseases but hugely dependent on the interpretational skill of an expert radiologist. We propose a Computer-Aided Detection model using Deep Convolutional Neural Networks to automatically detect TB from Montgomery County (MC) Tuberculosis radiographs. Our proposed model performed at 87.1% validation accuracy and evaluated using confusion matrix and accuracy as metrics.

Computer-aided detection in chest radiography based on artificial intelligence: a survey

As the most common examination tool in medical practice, chest radiography has important clinical value in the diagnosis of disease. Thus, the automatic detection of chest disease based on chest radiography has become one of the hot topics in medical imaging research. Based on the clinical applications, the study conducts a comprehensive survey on computer-aided detection (CAD) systems, and especially focuses on the artificial intelligence technology applied in chest radiography. The paper presents several common chest X-ray datasets and briefly introduces general image preprocessing procedures, such as contrast enhancement and segmentation, and bone suppression techniques that are applied to chest radiography. Then, the CAD system in the detection of specific disease (pulmonary nodules, tuberculosis, and interstitial lung diseases) and multiple diseases is described, focusing on the basic principles of the algorithm, the data used in the study, the evaluation measures, and the results. Finally, the paper summarizes the CAD system in chest radiography based on artificial intelligence and discusses the existing problems and trends.

Automated Chest X-Ray Screening: Can Lung Region Symmetry Help Detect Pulmonary Abnormalities?

Our primary motivator is the need for screening HIV+ populations in resource-constrained regions for exposure to Tuberculosis, using posteroanterior chest radiographs (CXRs). The proposed method is motivated by the observation that radiological examinations routinely conduct bilateral comparisons of the lung field. In addition, the abnormal CXRs tend to exhibit changes in the lung shape, size, and content (textures), and in overall, reflection symmetry between them. We analyze the lung region symmetry using multi-scale shape features, and edge plus texture features. Shape features exploit local and global representation of the lung regions, while edge and texture features take internal content, including spatial arrangements of the structures. For classification, we have performed voting-based combination of three different classifiers: Bayesian network, multilayer perception neural networks, and random forest. We have used three CXR benchmark collections made available by the U.S. National Library of Medicine and the National Institute of Tuberculosis and Respiratory Diseases, India, and have achieved a maximum abnormality detection accuracy (ACC) of 91.00% and area under the ROC curve (AUC) of 0.96. The proposed method outperforms the previously reported methods by more than 5% in ACC and 3% in AUC.

Digital CXR with computer aided diagnosis versus symptom screen to define presumptive tuberculosis among household contacts and impact on tuberculosis diagnosis

BACKGROUND

Household (HH) contact tracing is a strategy that targets high risk groups for TB. Symptom based screening is the standard used to identify HH contacts at risk for TB during HH contact tracing for TB. However, this strategy may be limited due to poor performance in predicting TB. The objective of this study was to compare CXR with Computer Aided Diagnosis (CAD) against symptom screen for defining presumptive TB and how TB detection changes with each method.

METHODS

Household contacts of consecutive index bacteriologically confirmed TB cases were visited by study teams and given TB/HIV education to raise awareness of the risk of TB following close contact with a TB patient. Contacts were encouraged to visit the health facility for screening; where symptoms history was obtained and opt out HIV testing was provided as part of the screening process. CXR was offered to all regardless of symptoms, followed by definitive sputum test with either Xpert MTB RIF or smear microscopy.

RESULTS

Among 919 HH contacts that presented for screening, 865 were screened with CXR and 464 (53.6%) had an abnormal CXR and the rest had a normal CXR. Among 444 HH contacts with valid sputum results, 274 (61.7%) were symptom screen positive and 255 (57.4%) had an abnormal CXR. Overall, TB was diagnosed in 32/444 (7.2%); 13 bacteriologically unconfirmed and 19 bacteriologically confirmed. Of 19 bacteriologically confirmed TB 8 (42.1%) were symptom screen negative contacts with an abnormal CXR and these 6/8 (75.0%) were HIV positive. Among the 13 bacteriologically unconfirmed TB cases, 7 (53.8%) were HIV positive and all had an abnormal CXR.

CONCLUSION

Symptom screen if used alone with follow on definitive TB testing only for symptom screen positive individuals would have missed eight of the 19 confirmed TB cases detected in this study. There is need to consider use of other screening strategies apart from symptom screen alone for optimal rule out of TB especially in HIV positive individuals that are at greatest risk of TB and present atypically.

Computer-Aided Diagnosis for Chest Radiographs in Intensive Care

The chest radiograph is an essential tool for the diagnosis of several lung diseases in intensive care units (ICU). However, several factors make the interpretation of the chest radiograph difficult including the number of X-rays done daily in ICU, the quality of the chest radiograph, and the lack of a standardized interpretation. To overcome these limitations in the interpretation of chest radiographs, researchers have developed computer-aided diagnosis (CAD) systems. In this review, the authors report the methodology used to develop CAD systems including identification of the region of interest, analysis of these regions, and classification. Currently, only a few CAD systems for chest X-ray interpretation are commercially available. Some promising research is ongoing, but the involvement of the pediatric research community is needed for the development and validation of such CAD systems dedicated to pediatric intensive care.

On Combining Multiple-Instance Learning and Active Learning for Computer-Aided Detection of Tuberculosis

The major advantage of multiple-instance learning (MIL) applied to a computer-aided detection (CAD) system is that it allows optimizing the latter with case-level labels instead of accurate lesion outlines as traditionally required for a supervised approach. As shown in previous work, a MIL-based CAD system can perform comparably to its supervised counterpart considering complex tasks such as chest radiograph scoring in tuberculosis (TB) detection. However, despite this remarkable achievement, the uncertainty inherent to MIL can lead to a less satisfactory outcome if analysis at lower levels (e.g., regions or pixels) is needed. This issue may seriously compromise the applicability of MIL to tasks related to quantification or grading, or detection of highly localized lesions. In this paper, we propose to reduce uncertainty by embedding a MIL classifier within an active learning (AL) framework. To minimize the labeling effort, we develop a novel instance selection mechanism that exploits the MIL problem definition through one-class classification. We adapt this mechanism to provide meaningful regions instead of individual instances for expert labeling, which is a more appropriate strategy given the application domain. In addition, and contrary to usual AL methods, a single iteration is performed. To show the effectiveness of our approach, we compare the output of a MIL-based CAD system trained with and without the proposed AL framework. The task is to detect textural abnormalities related to TB. Both quantitative and qualitative evaluations at the pixel level are carried out. Our method significantly improves the MIL-based classification.

Development of two artificial neural network models to support the diagnosis of pulmonary tuberculosis in hospitalized patients in Rio de Janeiro, Brazil

Pulmonary tuberculosis (PTB) remains a worldwide public health problem. Diagnostic algorithms to identify the best combination of diagnostic tests for PTB in each setting are needed for resource optimization. We developed one artificial neural network model for classification (multilayer perceptron-MLP) and another risk group assignment (self-organizing map-SOM) for PTB in hospitalized patients in a high complexity hospital in Rio de Janeiro City, using clinical and radiologic data collected from 315 presumed PTB cases admitted to isolation rooms from March 2003 to December 2004 (TB prevalence = 21.5 %). The MLP model included 7 variables-radiologic classification, age, gender, cough, night sweats, weight loss and anorexia. The sensitivity of the MLP model was 96.0 % (95 % CI ±2.0), the specificity was 89.0 % (95 % CI ±2.0), the positive predictive value was 72.5 % (95 % CI ±3.5) and the negative predictive value was 98.5 % (95 % CI ±0.5). The variable with the highest discriminative power was the radiologic classification. The high negative predictive value found in the MLP model suggests that the use of this model at the moment of hospital admission is safe. SOM model was able to correctly assign high-, medium- and low-risk groups to patients. If prospective validation in other series is confirmed, these models can become a tool for decision-making in tertiary health facilities in countries with limited resources.

Edge map analysis in chest X-rays for automatic pulmonary abnormality screening

PURPOSE

Our particular motivator is the need for screening HIV+ populations in resource-constrained regions for the evidence of tuberculosis, using posteroanterior chest radiographs (CXRs).

METHOD

The proposed method is motivated by the observation that abnormal CXRs tend to exhibit corrupted and/or deformed thoracic edge maps. We study histograms of thoracic edges for all possible orientations of gradients in the range [Formula: see text] at different numbers of bins and different pyramid levels, using five different regions-of-interest selection.

RESULTS

We have used two CXR benchmark collections made available by the U.S. National Library of Medicine and have achieved a maximum abnormality detection accuracy (ACC) of 86.36 % and area under the ROC curve (AUC) of 0.93 at 1 s per image, on average.

CONCLUSION

We have presented an automatic method for screening pulmonary abnormalities using thoracic edge map in CXR images. The proposed method outperforms previously reported state-of-the-art results.

Implementation Research to Inform the Use of Xpert MTB/RIF in Primary Health Care Facilities in High TB and HIV Settings in Resource Constrained Settings

Background

The current cost of Xpert MTB RIF (Xpert) consumables is such that algorithms are needed to select which patients to prioritise for testing with Xpert.

Objective

To evaluate two algorithms for prioritisation of Xpert in primary health care settings in a high TB and HIV burden setting.

Method

Consecutive, presumptive TB patients with a cough of any duration were offered either Xpert or Fluorescence microscopy (FM) test depending on their CXR score or HIV status. In one facility, sputa from patients with an abnormal CXR were tested with Xpert and those with a normal CXR were tested with FM (“CXR algorithm”). CXR was scored automatically using a Computer Aided Diagnosis (CAD) program. In the other facility, patients who were HIV positive were tested using Xpert and those who were HIV negative were tested with FM (“HIV algorithm”).

Results

Of 9482 individuals pre-screened with CXR, Xpert detected TB in 2090/6568 (31.8%) with an abnormal CXR, and FM was AFB positive in 8/2455 (0.3%) with a normal CXR. Of 4444 pre-screened with HIV, Xpert detected TB in 508/2265 (22.4%) HIV positive and FM was AFB positive in 212/1920 (11.0%) in HIV negative individuals. The notification rate of new bacteriologically confirmed TB increased; from 366 to 620/ 100,000/yr and from 145 to 261/100,000/yr at the CXR and HIV algorithm sites respectively. The median time to starting TB treatment at the CXR site compared to the HIV algorithm site was; 1(IQR 1-3 days) and 3 (2-5 days) (p<0.0001) respectively.

Conclusion

Use of Xpert in a resource-limited setting at primary care level in conjunction with pre-screening tests reduced the number of Xpert tests performed. The routine use of Xpert resulted in additional cases of confirmed TB patients starting treatment. However, there was no increase in absolute numbers of patients starting TB treatment. Same day diagnosis and treatment commencement was achieved for both bacteriologically confirmed and empirically diagnosed patients where Xpert was used in conjunction with CXR.

Cavity contour segmentation in chest radiographs using supervised learning and dynamic programming

PURPOSE

Efficacy of tuberculosis (TB) treatment is often monitored using chest radiography. Monitoring size of cavities in pulmonary tuberculosis is important as the size predicts severity of the disease and its persistence under therapy predicts relapse. The authors present a method for automatic cavity segmentation in chest radiographs.

METHODS

A two stage method is proposed to segment the cavity borders, given a user defined seed point close to the center of the cavity. First, a supervised learning approach is employed to train a pixel classifier using texture and radial features to identify the border pixels of the cavity. A likelihood value of belonging to the cavity border is assigned to each pixel by the classifier. The authors experimented with four different classifiers:k-nearest neighbor (kNN), linear discriminant analysis (LDA), GentleBoost (GB), and random forest (RF). Next, the constructed likelihood map was used as an input cost image in the polar transformed image space for dynamic programming to trace the optimal maximum cost path. This constructed path corresponds to the segmented cavity contour in image space.

RESULTS

The method was evaluated on 100 chest radiographs (CXRs) containing 126 cavities. The reference segmentation was manually delineated by an experienced chest radiologist. An independent observer (a chest radiologist) also delineated all cavities to estimate interobserver variability. Jaccard overlap measure Ω was computed between the reference segmentation and the automatic segmentation; and between the reference segmentation and the independent observer's segmentation for all cavities. A median overlap Ω of 0.81 (0.76 ± 0.16), and 0.85 (0.82 ± 0.11) was achieved between the reference segmentation and the automatic segmentation, and between the segmentations by the two radiologists, respectively. The best reported mean contour distance and Hausdorff distance between the reference and the automatic segmentation were, respectively, 2.48 ± 2.19 and 8.32 ± 5.66 mm, whereas these distances were 1.66 ± 1.29 and 5.75 ± 4.88 mm between the segmentations by the reference reader and the independent observer, respectively. The automatic segmentations were also visually assessed by two trained CXR readers as "excellent," "adequate," or "insufficient." The readers had good agreement in assessing the cavity outlines and 84% of the segmentations were rated as "excellent" or "adequate" by both readers.

CONCLUSIONS

The proposed cavity segmentation technique produced results with a good degree of overlap with manual expert segmentations. The evaluation measures demonstrated that the results approached the results of the experienced chest radiologists, in terms of overlap measure and contour distance measures. Automatic cavity segmentation can be employed in TB clinics for treatment monitoring, especially in resource limited settings where radiologists are not available.

A novel multiple-instance learning-based approach to computer-aided detection of tuberculosis on chest X-rays

To reach performance levels comparable to human experts, computer-aided detection (CAD) systems are typically optimized following a supervised learning approach that relies on large training databases comprising manually annotated lesions. However, manually outlining those lesions constitutes a difficult and time-consuming process that renders detailedly annotated data difficult to obtain. In this paper, we investigate an alternative approach, namely multiple-instance learning (MIL), that does not require detailed information for optimization. We have applied MIL to a CAD system for tuberculosis detection. Only the case condition (normal or abnormal) was required during training. Based upon the well-known miSVM technique, we propose an improved algorithm that overcomes miSVM's drawbacks related to positive instance underestimation and costly iteration. To show the advantages of our MIL-based approach as compared with a traditional supervised one, experiments with three X-ray databases were conducted. The area under the receiver operating characteristic curve was utilized as a performance measure. With the first database, for which training lesion annotations were available, our MIL-based method was comparable to the supervised system ( 0.86 versus 0.88 ). When evaluating the remaining databases, given their large difference with the previous image set, the most appealing strategy was to retrain the CAD systems. However, since only the case condition was available, only the MIL-based system could be retrained. This scenario, which is common in real-world applications, demonstrates the better adaptation capabilities of the proposed approach. After retraining, our MIL-based system significantly outperformed the supervised one ( 0.86 versus 0.79 and 0.91 versus 0.85 , and p=0.0002 , respectively).

Role of Gist and PHOG features in computer-aided diagnosis of tuberculosis without segmentation

Purpose

Effective diagnosis of tuberculosis (TB) relies on accurate interpretation of radiological patterns found in a chest radiograph (CXR). Lack of skilled radiologists and other resources, especially in developing countries, hinders its efficient diagnosis. Computer-aided diagnosis (CAD) methods provide second opinion to the radiologists for their findings and thereby assist in better diagnosis of cancer and other diseases including TB. However, existing CAD methods for TB are based on the extraction of textural features from manually or semi-automatically segmented CXRs. These methods are prone to errors and cannot be implemented in X-ray machines for automated classification.

Methods

Gabor, Gist, histogram of oriented gradients (HOG), and pyramid histogram of oriented gradients (PHOG) features extracted from the whole image can be implemented into existing X-ray machines to discriminate between TB and non-TB CXRs in an automated manner. Localized features were extracted for the above methods using various parameters, such as frequency range, blocks and region of interest. The performance of these features was evaluated against textural features. Two digital CXR image datasets (8-bit DA and 14-bit DB) were used for evaluating the performance of these features.

Results

Gist (accuracy 94.2% for DA, 86.0% for DB) and PHOG (accuracy 92.3% for DA, 92.0% for DB) features provided better results for both the datasets. These features were implemented to develop a MATLAB toolbox, TB-Xpredict, which is freely available for academic use at http://sourceforge.net/projects/tbxpredict/. This toolbox provides both automated training and prediction modules and does not require expertise in image processing for operation.

Conclusion

Since the features used in TB-Xpredict do not require segmentation, the toolbox can easily be implemented in X-ray machines. This toolbox can effectively be used for the mass screening of TB in high-burden areas with improved efficiency.

Diagnostic accuracy of computer-aided detection of pulmonary tuberculosis in chest radiographs: a validation study from sub-Saharan Africa

Background

Chest radiography to diagnose and screen for pulmonary tuberculosis has limitations, especially due to inter-reader variability. Automating the interpretation has the potential to overcome this drawback and to deliver objective and reproducible results. The CAD4TB software is a computer-aided detection system that has shown promising preliminary findings. Evaluation studies in different settings are needed to assess diagnostic accuracy and practicability of use.

Methods

CAD4TB was evaluated on chest radiographs of patients with symptoms suggestive of pulmonary tuberculosis enrolled in two cohort studies in Tanzania. All patients were characterized by sputum smear microscopy and culture including subsequent antigen or molecular confirmation of Mycobacterium tuberculosis (M.tb) to determine the reference standard. Chest radiographs were read by the software and two human readers, one expert reader and one clinical officer. The sensitivity and specificity of CAD4TB was depicted using receiver operating characteristic (ROC) curves, the area under the curve calculated and the performance of the software compared to the results of human readers.

Results

Of 861 study participants, 194 (23%) were culture-positive for M.tb. The area under the ROC curve of CAD4TB for the detection of culture-positive pulmonary tuberculosis was 0.84 (95% CI 0.80–0.88). CAD4TB was significantly more accurate for the discrimination of smear-positive cases against non TB patients than for smear-negative cases (p-value<0.01). It differentiated better between TB cases and non TB patients among HIV-negative compared to HIV-positive individuals (p<0.01). CAD4TB significantly outperformed the clinical officer, but did not reach the accuracy of the expert reader (p = 0.02), for a tuberculosis specific reading threshold.

Conclusion

CAD4TB accurately distinguished between the chest radiographs of culture-positive TB cases and controls. Further studies on cost-effectiveness, operational and ethical aspects should determine its place in diagnostic and screening algorithms.

The sensitivity and specificity of using a computer aided diagnosis program for automatically scoring chest X-rays of presumptive TB patients compared with Xpert MTB/RIF in Lusaka Zambia

Objective

To determine the sensitivity and specificity of a Computer Aided Diagnosis (CAD) program for scoring chest x-rays (CXRs) of presumptive tuberculosis (TB) patients compared to Xpert MTB/RIF (Xpert).

Method

Consecutive presumptive TB patients with a cough of any duration were offered digital CXR, and opt out HIV testing. CXRs were electronically scored as normal (CAD score ≤60) or abnormal (CAD score>60) using a CAD program. All patients regardless of CAD score were requested to submit a spot sputum sample for testing with Xpert and a spot and morning sample for testing with LED Fluorescence Microscopy-(FM).

Results

Of 350 patients with evaluable data, 291 (83.1%) had an abnormal CXR score by CAD. The sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of CXR compared to Xpert were 100% (95%CI 96.2–100), 23.2% (95%CI 18.2–28.9), 33.0% (95%CI 27.6–38.7) and 100% (95% 93.9–100), respectively. The area under the receiver operator curve (AUC) for CAD was 0.71 (95%CI 0.66–0.77). CXR abnormality correlated with smear grade (r = 0.30, p<0.0001) and with Xpert C_T(r = 0.37, p<0.0001).

Conclusions

To our knowledge this is the first time that a CAD program for TB has been successfully tested in a real world setting. The study shows that the CAD program had high sensitivity but low specificity and PPV. The use of CAD with digital CXR has the potential to increase the use and availability of chest radiography in screening for TB where trained human resources are scarce.

Tuberculosis screening in high human immunodeficiency virus prevalence settings: turning promise into reality

Twenty years of sky-high tuberculosis (TB) incidence rates and high TB mortality in high human immunodeficiency virus (HIV) prevalence countries have so far not been matched by the same magnitude or breadth of responses as seen in malaria or HIV programmes. Instead, recommendations have been narrowly focused on people presenting to health facilities for investigation of TB symptoms, or for HIV testing and care. However, despite the recent major investment and scale-up of TB and HIV services, undiagnosed TB remains highly prevalent at community level, implying that diagnosis of TB remains slow and incomplete. This maintains high transmission rates and exposes people living with HIV to high rates of morbidity and mortality. More intensive use of TB screening, with broader definitions of target populations, expanded indications for screening both inside and outside of health facilities, and appropriate selection of new diagnostic tools, offers the prospect of rapidly improving population-level control of TB. Diagnostic accuracy of suitable (high throughput) algorithms remains the major barrier to realising this goal. In the present study, we review the evidence available to guide expanded TB screening in HIV-prevalent settings, ideally through combined TB-HIV interventions that provide screening for both TB and HIV, and maximise entry to HIV and TB care and prevention. Ideally, we would systematically test, treat and prevent TB and HIV comprehensively, offering both TB and HIV screening to all health facility attendees, TB households and all adults in the highest risk communities. However, we are still held back by inadequate diagnostics, financing and paucity of population-impact data. Relevant contemporary research showing the high need for potential gains, and pitfalls from expanded and intensified TB screening in high HIV prevalence settings are discussed in this review.

Automatic tuberculosis screening using chest radiographs

Tuberculosis is a major health threat in many regions of the world. Opportunistic infections in immunocompromised HIV/AIDS patients and multi-drug-resistant bacterial strains have exacerbated the problem, while diagnosing tuberculosis still remains a challenge. When left undiagnosed and thus untreated, mortality rates of patients with tuberculosis are high. Standard diagnostics still rely on methods developed in the last century. They are slow and often unreliable. In an effort to reduce the burden of the disease, this paper presents our automated approach for detecting tuberculosis in conventional posteroanterior chest radiographs. We first extract the lung region using a graph cut segmentation method. For this lung region, we compute a set of texture and shape features, which enable the X-rays to be classified as normal or abnormal using a binary classifier. We measure the performance of our system on two datasets: a set collected by the tuberculosis control program of our local county's health department in the United States, and a set collected by Shenzhen Hospital, China. The proposed computer-aided diagnostic system for TB screening, which is ready for field deployment, achieves a performance that approaches the performance of human experts. We achieve an area under the ROC curve (AUC) of 87% (78.3% accuracy) for the first set, and an AUC of 90% (84% accuracy) for the second set. For the first set, we compare our system performance with the performance of radiologists. When trying not to miss any positive cases, radiologists achieve an accuracy of about 82% on this set, and their false positive rate is about half of our system's rate.

Optimal triage test characteristics to improve the cost-effectiveness of the Xpert MTB/RIF assay for TB diagnosis: a decision analysis

Background

High costs are a limitation to scaling up the Xpert MTB/RIF assay (Xpert) for the diagnosis of tuberculosis in resource-constrained settings. A triaging strategy in which a sensitive but not necessarily highly specific rapid test is used to select patients for Xpert may result in a more affordable diagnostic algorithm. To inform the selection and development of particular diagnostics as a triage test we explored combinations of sensitivity, specificity and cost at which a hypothetical triage test will improve affordability of the Xpert assay.

Methods

In a decision analytical model parameterized for Uganda, India and South Africa, we compared a diagnostic algorithm in which a cohort of patients with presumptive TB received Xpert to a triage algorithm whereby only those with a positive triage test were tested by Xpert.

Findings

A triage test with sensitivity equal to Xpert, 75% specificity, and costs of US$5 per patient tested reduced total diagnostic costs by 42% in the Uganda setting, and by 34% and 39% respectively in the India and South Africa settings. When exploring triage algorithms with lower sensitivity, the use of an example triage test with 95% sensitivity relative to Xpert, 75% specificity and test costs $5 resulted in similar cost reduction, and was cost-effective by the WHO willingness-to-pay threshold compared to Xpert for all in Uganda, but not in India and South Africa. The gain in affordability of the examined triage algorithms increased with decreasing prevalence of tuberculosis among the cohort.

Conclusions

A triage test strategy could potentially improve the affordability of Xpert for TB diagnosis, particularly in low-income countries and with enhanced case-finding. Tests and markers with lower accuracy than desired of a diagnostic test may fall within the ranges of sensitivity, specificity and cost required for triage tests and be developed as such.

Foreign object detection and removal to improve automated analysis of chest radiographs

PURPOSE

Chest radiographs commonly contain projections of foreign objects, such as buttons, brassier clips, jewellery, or pacemakers and wires. The presence of these structures can substantially affect the output of computer analysis of these images. An automated method is presented to detect, segment, and remove foreign objects from chest radiographs.

METHODS

Detection is performed using supervised pixel classification with a kNN classifier, resulting in a probability estimate per pixel to belong to a projected foreign object. Segmentation is performed by grouping and post-processing pixels with a probability above a certain threshold. Next, the objects are replaced by texture inpainting.

RESULTS

The method is evaluated in experiments on 257 chest radiographs. The detection at pixel level is evaluated with receiver operating characteristic analysis on pixels within the unobscured lung fields and an Az value of 0.949 is achieved. Free response operator characteristic analysis is performed at the object level, and 95.6% of objects are detected with on average 0.25 false positive detections per image. To investigate the effect of removing the detected objects through inpainting, a texture analysis system for tuberculosis detection is applied to images with and without pathology and with and without foreign object removal. Unprocessed, the texture analysis abnormality score of normal images with foreign objects is comparable to those with pathology. After removing foreign objects, the texture score of normal images with and without foreign objects is similar, while abnormal images, whether they contain foreign objects or not, achieve on average higher scores.

CONCLUSIONS

The authors conclude that removal of foreign objects from chest radiographs is feasible and beneficial for automated image analysis.

Detecting tuberculosis in radiographs using combined lung masks

Tuberculosis (TB) is a major health threat in many regions of the world, while diagnosing tuberculosis still remains a challenge. Mortality rates of patients with undiagnosed TB are high. Modern diagnostic techniques are often too slow or too expensive for highly-populated developing countries that bear the brunt of the disease. In an effort to reduce the burden of the disease, this paper presents an automated approach for detecting TB on conventional posteroanterior chest radiographs. The idea is to provide developing countries, which have limited access to radiological services and radiological expertise, with an inexpensive detection system that allows screening of large parts of the population in rural areas. In this paper, we present results produced by our TB screening system. We combine a lung shape model, a segmentation mask, and a simple intensity model to achieve a better segmentation mask for the lung. With the improved masks, we achieve an area under the ROC curve of more than 83%, measured on data compiled within a tuberculosis control program.