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.
Highlights
Tuberculosis (TB) is a contagious disease that is designated among 10 highest causes of death and the leading infectious disease above than human immunodeficiency Virus (HIV)/ acquired immune deficiency syndrome (AIDS) [1]
By applying B-convolutional neural network (CNN), TB identification task can be described as follows; consider a set of chest x-rays (CXRs) denoted as I = {I1,I2, . . . ,IN }, where each CXR is described as CXR I ∈ [0; 1]m×n (m is the height and n is the width of all the CXRs denoted by N number of CXRs) as a set of parallel labels Y = {y1,y2, . . . ,yN } where all the labels corresponds to a binary classification result U ∈ [0; 1]
Proposed based convolutional neural network (B-CNN) exploits the model uncertainty and Bayesian confidence to improve the accuracy of TB identification as well as validation of the results
Summary
Tuberculosis (TB) is a contagious disease that is designated among 10 highest causes of death and the leading infectious disease above than human immunodeficiency Virus (HIV)/ acquired immune deficiency syndrome (AIDS) [1]. Each year millions of people get infected with tuberculosis. In 2017, approximately 1.3 million deaths were recorded for HIV-negative people. The top estimation was that in total 10 million individuals suffered from TB infection disease in 2017. According to the World Health Organization (WHO), the affected tuberculosis individuals were from all countries and age groups, i.e., over 90% were adults whereas, 9% of people were living with HIV. WHO 2018 TB report listed 30 high TB burden countries from which eight highest TB affected countries were: India (27%), China (9%), Indonesia (8%), Philippines (6%), Pakistan (5%), Nigeria (4%), Bangladesh (4%) and
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
