Role of Artificial Intelligence in Diagnosis of Tuberculosis: An Investigation
AUTHORS
Avree Ito-Fujita,Nancy Atmospera-Walch School of Nursing, University of Hawaii at Mānoa, Honolulu, USA
Shayna Katz,Nancy Atmospera-Walch School of Nursing, University of Hawaii at Mānoa, Honolulu, USA
ABSTRACT
Tuberculosis (TB) is a worldwide problem that poses a severe threat to public health. Pathology is among the most essential methods for diagnosing tuberculosis in clinical practice. Identifying specifically colored tuberculosis bacilli under a microscope is essential for confirming tuberculosis as the diagnosis. Even for expert pathologists, it is time-consuming and laborious due to the tiny size and quantity of bacilli. Then, this strenuosity often results in poor detection rates and incorrect diagnoses. TB is the most contagious human disease, driven by the bacteria Mycobacterium Tuberculosis (MTB). It is an infectious illness that spreads from one person to another via the air. Pulmonary TB (PTB) is a tuberculosis that primarily affects the lungs. It may, however, affect other body regions, including the brain, bones, and lymph nodes. It is also known as Extrapulmonary Tuberculosis (EPTB). Because tuberculosis has common symptoms, it may be difficult to tell whether a patient has it without adequate testing. Employing image analysis as well as Artificial Intelligence-based classification techniques, an accurate and unique system for detecting tuberculosis has been developed in this work. There are two stages to the intended system. Three different Artificial Intelligence-based classification algorithms are applied once the X-ray image is initially processed utilizing pretreatment, classification, and feature extraction.
KEYWORDS
Tuberculosis, Artificial intelligence, Mycobacterium tuberculosis, Machine learning, Deep learning
REFERENCES
[1] G. Chartrand, P. M. Cheng, and E. Vorontsov, “Deep learning: a primer for radiologists,” Radiographics, vol.37, pp. 2113-2131, (2017)
[2] A. Krizhevsky, I. Sutskever, and G. E. Hinton, “Imagenet classification with deep convolutional neural networks,” Adv Neural Inf Process Syst, pp.1097-1105, (2012)
[3] WHO, Global Tuberculosis Report 2020. World Health Organization, Geneva, (2020)
[4] R. R. Nathavitharana, C. Yoon, and P. Macpherson, “Guidance for studies evaluating the accuracy of tuberculosis triage tests,” J Infect Dis.; vol.220, pp.116-125, (2019)
[5] M. Rahman, M. S. Flora, and M. M. Husain, “National tuberculosis prevalence survey 2015–2016,” Ministry of Health and Family Welfare, Dhaka, (2016)
[6] Goodfellow I, Bengio Y, Courville A, Deep learning. MIT Press, Cambridge, MA, (2016)
[7] Z. Z. Qin, T. Naheyan, M. Ruhwald, “A new resource on artificial intelligence-powered computer-automated detection software products for tuberculosis programmers and implementers,” Tuberculosis (Edinb), 127102049, (2021)
[8] The Stop TB Partnership, FIND. Resource center on computer-aided detection products for the diagnosis of tuberculosis. https://www.ai4hlth.org/ Date: 2020 Date accessed: September 10, (2020)
[9] WHO, WHO consolidated guidelines on tuberculosis: module 2: Screening: systematic screening for tuberculosis disease. World Health Organization, Geneva, (2021)
[10] K. Murphy, S. S. Habib, and S. M. A. Zaidi, “Computer-aided detection of tuberculosis on chest radiographs: an evaluation of the CAD4TB v6 system,” Sci Rep., 105492, (2020)
[11] S. M. A. Zaidi, S. S. Habib, and B. Van Ginneken, “Evaluation of the diagnostic accuracy of computer-aided detection of tuberculosis on chest radiography among private sector patients in Pakistan,” Sci Rep.; 812339, (2018)
[12] T. Pande, C. Cohen, M. Pai, and F. Ahmad Khan, “Computer-aided detection of pulmonary tuberculosis on digital chest radiographs: A systematic review,” Int J Tuberc Lung Dis.; vol.20, pp.1226-1230, (2016)
[13] D. B. Larson, H. Harvey, D. L. Rubin, N. Irani, R. T. Justin, and C. P. Langlotz, “Regulatory frameworks for development and evaluation of artificial intelligence-based diagnostic imaging algorithms: Summary and recommendations,” J Am Coll Radiol., vol.18, pp.413-424, (2021)
[14] M. Nash, R. Kadavigere, and J. Andrade, “Deep learning, computer-aided radiography reading for tuberculosis: A diagnostic accuracy study from a tertiary hospital in India. Sci Rep.; vol.10, pp.210, (2020)
[15] Z. Z. Qin, M. S. Sander, and B. Rai, “Using artificial intelligence to read chest radiographs for tuberculosis detection: A multi-site evaluation of the diagnostic accuracy of three deep learning systems,” Sci Rep., 915000, (2019)
[16] A. P. Bradley, “The use of the area under the ROC curve in the evaluation of machine learning algorithms,” Pattern Recognit., vol.30, pp.1145-1159, (1997)
[17] P. M. Bossuyt, J. B. Reitsma, and D. E. Bruns, “Towards complete and accurate reporting of studies of diagnostic accuracy: The STARD initiative,” Croat Med J., vol.44, pp.635-638, (2003)
[18] S. Banu, F. Haque, and S. Ahmed, “Social Enterprise Model (SEM) for private sector tuberculosis screening and care in Bangladesh,” PLoS One, 15e0241437, (2020)
[19] S. M. McKinney, M. Sieniek, and V. Godbole, “International evaluation of an AI system for breast cancer screening. Nature, vol.577, pp.89-94. (2020)
[20] T. Saito and M. Rehmsmeier, “The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets,” PLoS One., 10e0118432, (2015)