Uncertainty Assisted Robust Tuberculosis Identification With Bayesian Convolutional Neural Networks

Abstract

Tuberculosis (TB) is an infectious disease that can lead towards death if left untreated. TB detection involves extraction of complex TB manifestation features such as lung cavity, air space consolidation, endobronchial spread, and pleural effusions from chest x-rays (CXRs). Deep learning based approach named convolutional neural network (CNN) has the ability to learn complex features from CXR images. The main problem is that CNN does not consider uncertainty to classify CXRs using softmax layer. It lacks in presenting the true probability of CXRs by differentiating confusing cases during TB detection. This paper presents the solution for TB identification by using Bayesian-based convolutional neural network (B-CNN). It deals with the uncertain cases that have low discernibility among the TB and non-TB manifested CXRs. The proposed TB identification methodology based on B-CNN is evaluated on two TB benchmark datasets, i.e., Montgomery and Shenzhen. For training and testing of proposed scheme we have utilized Google Colab platform which provides NVidia Tesla K80 with 12 GB of VRAM, single core of 2.3 GHz Xeon Processor, 12 GB RAM and 320 GB of disk. B-CNN achieves 96.42% and 86.46% accuracy on both dataset, respectively as compared to the state-of-the-art machine learning and CNN approaches. Moreover, B-CNN validates its results by filtering the CXRs as confusion cases where the variance of B-CNN predicted outputs is more than a certain threshold. Results prove the supremacy of B-CNN for the identification of TB and non-TB sample CXRs as compared to counterparts in terms of accuracy, variance in the predicted probabilities and model uncertainty. INDEX TERMS Tuberculosis identification, computer-aided diagnostics, medical image analysis, Bayesian convolutional neural networks, model uncertainty.

Lung Segmentation-Based Pulmonary Disease Classification Using Deep Neural Networks

“…Besides the above-discussed shape and texture featurebased techniques, CNN based advanced methods are also proposed in the literature for the disease classification from the entire CXR. In this regard, numerous techniques have been proposed for tuberculosis classification [28], [29]. The authors in [28] utilized the pre-trained CNN architectures (AlexNet and GoogleNet) for disease classification and obtained optimized results when compared with the existing shape and texture-based techniques.…”

Systematic Study on Diagnosis of Lung Disorders using Machine Learning and Deep Learning Algorithms

“…B-CNN (CNN with Bayesian Optimization Algorithm) validates results produced by the SoftMax layer by filtering confusion cases on calculating variance to increase accuracy. BOA works on the Bayesian theorem (i.e., calculates the probability of ongoing event) and tedious [14,24].…”

“…Zain Ul Abideen et al, predict tuberculosis by implementing B-CNN. The images from Montgomery and Shenzhen are taken and fed to the CNN model [24]. The performance of CNN fails when uncertain cases occur.…”

A Survey of Deep Learning for Lung Disease Detection on Medical Images: State-of-the-Art, Taxonomy, Issues and Future Directions

“…CNN has also been used to classify tuberculosis [ 65 , 66 , 67 ]. Ul Abideen et al [ 68 ] used a Bayesian-based CNN that exploits the model uncertainty and Bayesian confidence to improve the accuracy of tuberculosis identification. In other work, a deep CNN algorithm named deep learning-based automatic detection (DLAD), was developed for tuberculosis classification that contains 27 layers with 12 residual connections [ 69 ].…”