Aims and objectives
Tuberculosis (TB) is a leading cause of death worldwide and while there are effective ways to prevent,
detect and cure the disease,
it remains a significant global public health challenge that has sparked the concern of the World Health Organization [5].
The use of chest radiographs (CXR) has long been part of the triaging,
screening and diagnostic strategy for TB [6].
However,
despite improved screening methods,
detection technologies and curative treatment,
the burden of TB remains a huge issue.
While the development of advanced modern...
Methods and materials
Using a keyphrase search approach,
a set of RIS search queries were created to identify chest radiograph studies highly suspicious for tuberculosis.
Frontal view DICOM images were exported.
Using this search,
205 frontal images of radiographically suspicious tuberculosis were identified,
with 301 of 820 quadrants containing disease.
8,933 normal chest radiographs were used as negative control.
We employ a dual-channel ImageNet pre-trained deep convolutional neural network (DCNN) derived from the Inception-v2-Resnet architecture to detect tuberculosis (Figure 1).
The dataset was split using 70% training,
10%...
Results
Objective Network Performance Evaluation:
Model performance was assessed by independently assessing the AUC of both the (a) Full View CNN as well as the (b) Quad View CNN. The Full View model obtains an AUC of 0.941 on the test set after training for 162 epochs with a test accuracy of 0.953 (Figure 2a).
The Quadrant View model obtains an AUC of 0.819 after training for 37 epochs with a test accuracy of 0.900 (Figure 2b).
Subjective Network Evaluation:
Class-activation map analysis demonstrated stronger activation...
Conclusion
Comparison of Network Performance:
Our trained neural network yielded an AUC of 0.82-0.94 which is comparable to previously published detection statistics of other experimental neural networks.
For instance,
Hwang et al.
(2016) had AUCs ranging from 0.88 to 0.96 [2].
Our study shows that DCNN transfer learning approach is able to identify TB on chest radiographs.
Subjective Evaluations Patterns:
However,
unlike previous studies,
we wish to draw certain additional findings,
particularly from the subjective network evaluation.
From our analyses with the generated test Class Activation...
References
1.
Ahmad Khan F,
Pande T,
Tessema B,
et al.
Computer-aided reading of tuberculosis chest radiography: moving the research agenda forward to inform policy.
Eur Respir J 2017; 50: 1700953.
doi: 10.1183/13993003.00953-2017.
2.
Hwang S,
Kim HE,
Jeong J,
Kim HJ.
A novel approach for tuberculosis screening based on deep convolutional neural networks.
In: Tourassi GD,
Armato SG,
eds.
Proceedings of SPIE: medical imaging 2016—title.
Vol 9785.
Bellingham,
Wash: International Society for Optics and Photonics,
2016; 97852W.
3.
Lakhani P,
Sundaram B.
Deep Learning at...