Automatic detection of tuberculosis bacilli from microscopic sputum smear images using deep learning methods

• Autorzy: Panicker R. O. , Kalmady K. S. , Rajan J. , Sabu M. K.

Identyfikatory

DOI

10.1016/j.bbe.2018.05.007

• Języki publikacji: EN

Abstrakty

EN

An automatic method for the detection of Tuberculosis (TB) bacilli from microscopic sputum smear images is presented in this paper. According to WHO, TB is the ninth leading cause of death all over the world. There are various techniques to diagnose TB, of which conventional microscopic sputum smear examination is considered to be the gold standard. However, the aforementioned method of diagnosis is time intensive and error prone, even in experienced hands. The proposed method performs detection of TB, by image binarization and subsequent classification of detected regions using a convolutional neural network. We have evaluated our algorithm using a dataset of 22 sputum smear microscopic images with different backgrounds (high density and low-density images). Experimental results show that the proposed algorithm achieves 97.13% recall, 78.4% precision and 86.76% F-score for the TB detection. The proposed method automatically detects whether the sputum smear images is infected with TB or not. This method will aid clinicians to predict the disease accurately in a short span of time, thereby helping in improving the clinical outcome.

Słowa kluczowe

EN

tuberculosis bacilli; sputum smear microscopy; automatic TB detection; convolutional neural network; image processing

PL

prątek gruźlicy; mikroskopia rozmazu; wykrywanie gruźlicy; sieć neuronowa konwolucyjna; przetwarzanie obrazu

• Wydawca: Nałęcz Institute of Biocybernetics and Biomedical Engineering of the Polish Academy of Sciences ; Elsevier
• Czasopismo: Biocybernetics and Biomedical Engineering
• Rocznik: 2018
• Tom: Vol. 38, no. 3
• Strony: 691--699
• Opis fizyczny: Bibliogr. 32 poz., rys., tab., wykr.

Twórcy

autor

Panicker R. O.

• Department of Information Technology, College of Engineering, Trikaripur, Kerala, India; Department of Computer Applications, Cochin University of Science and Technology, Kochi, India

autor

Kalmady K. S.

• Department of Computer Science and Engineering, National Institute of Technology Karnataka, Surathkal, India

autor

Rajan J.

• Department of Computer Science and Engineering, National Institute of Technology Karnataka, Surathkal, India

autor

Sabu M. K.

• Department of Computer Applications, Cochin University of Science and Technology, Kochi, India

Bibliografia

• [1] Global tuberculosis report 2017, http://www.who.int/tb/publications/global_report/en/ [accessed 13.11.17].
• [2] Priya E, Srinivasan S. Separation of overlapping bacilli in microscopic digital TB images. Biocybern Biomed Eng 2015;35:87–99.
• [3] Panicker RO, Soman B, Saini G, Rajan J. A review of automatic methods based on image processing techniques for tuberculosis detection from microscopic sputum smear images. J Med Syst 2016;40(1):1–13.
• [4] Chang J, Arbeláez P, Switz N, Reber C, Tapley A, Davis JL, et al. Automated tuberculosis diagnosis using fluorescence images from a mobile microscope. Proc. of Medical Image Computing and Computer-Assisted Intervention – MICCAI. 7512. 2012. pp. 345–52.
• [5] Costa MG, Costa Filho CFF, Sena JF, Salem J, Lima MO. Automatic identification of Mycobacterium tuberculosis with conventional light microscopy. Proc. of 30th Annual International IEEE Eng Med Biol Soc; 2008. pp. 382–5.
• [6] Sotaquir'a M, Rueda L, Narvaez R. Detection and quantification of bacilli and clusters present in sputum smear samples: a novel algorithm for pulmonary tuberculosis diagnosis. Proc. of International Conference on Digital Image Processing; 2009. pp. 117–21.
• [7] Veropoulos K, Campbell C, Learmonth G, Knight B, Simpson J. The automated identification of tubercle bacilli using image processing and neural computing techniques. Proc. of the 8th International Conference on Artificial Neural Networks; 1998. pp. 797–802.
• [8] Veropoulos K, Campbell C, Learmonth G. Image processing and neural computing used in the diagnosis of tuberculosis. Proc. of IEE Colloquium on Intelligent Methods in Healthcare and Medical Applications (Digest No. 1998/514); 1998. 8/1-8/4.
• [9] Khutlang R, Krishnan S, Whitelaw A, Douglas TS. Automated detection of tuberculosis in Ziehl-Neelsen stained sputum smears using two one-class classifiers. J Microsc 2010;237(1):96–102.
• [10] Panicker RO, Soman B, Gangadharan KV, Sobhana NV. An adoption model describing clinician's acceptance of automated diagnostic system for tuberculosis. Health Technol 2016;10:1–11.
• [11] Sadaphal P, Rao J, Comstock GW, Beg MF. Image processing techniques for identifying Mycobacterium tuberculosis in Ziehl-Neelsen stains. Int J Tuberc Lung Dis 2008;12(5):579–82.
• [12] Makkapati V, Agrawal R, Acharya R. Segmentation and classification of tuberculosis bacilli from ZN-stained sputum smearimages. Proc. of IEEE International Conference on Automation Science and Engineering (CASE); 2009. pp. 217–20.
• [13] Sotaquirá M, Rueda L, Narvaez R. Detection and quantification of bacilli and clusters present in sputum smear samples: a novel algorithm for pulmonary tuberculosis diagnosis. Proc. of International Conference on Digital Image Processing; 2009. pp. 117–21.
• [14] Zhai Y, Liu Y, Zhou D, Liu S. Automatic Identification of Mycobacterium tuberculosis from ZN-stained sputum smear: algorithm and system design. Proc. of IEEE International Conference on Robotics and Biomimetics (ROBIO); 2010. pp. 41–6.
• [15] Nayak R, Shenoy VP, Galigekere RR. A new algorithm for automatic assessment of the degree of TB-infection using images of ZN-stained sputum smear. Proc. of International Conference on Systems in Medicine and Biology (ICSMB); 2010. pp. 294–9.
• [16] Khutlang R, Krishnan S, Dendere R, Whitelaw A, Veropoulos K, Learmonth G, et al. Classification of Mycobacterium tuberculosis in images of ZN stained sputum smears. IEEE Trans Inf Technol Biomed 2010;14(4):949–57.
• [17] Costa Filho CFF, Costa MGF. Sputum Smear Microscopy for Tuberculosis: Evaluation of Autofocus Functions and Automatic Identification of Tuberculosis Mycobacterium. Understanding Tuberculosis – Global Experiences and Innovative Approaches to the Diagnosis, Dr. Pere-Joan Cardona (Ed.), ISBN: 978-953-307-938-7, InTech, pp. 277–292, 2012.
• [18] Ayas S, Ekinci M. Random forest-based tuberculosis bacteria classification in images of ZN-stained sputum smear samples. SIViP 2010;8(1):49–61.
• [19] Osman MK, Ahmad F, Saad Z, Mashor MY, Jaafar H. A genetic algorithm-neural network approach for Mycobacterium tuberculosis detection in Ziehl-Neelsen stained tissue slide images. Proc. of 10th International Conference on Intelligent Systems Design and Applications (ISDA). 2010. pp. 1229–34.
• [20] Costa Filho CFF, Levy PC, Xavier CM, Fujimoto LBM, Costa MGF. Automatic identification of tuberculosis mycobacterium. Res Biomed Eng 2015;31(1):33–43.
• [21] Chao Xu, Zhou D, Liu Y. Segmentation of touching Mycobacterium tuberculosis from Ziehl-Neelsen stained sputum smear images. Proc. of SPIE vol. 9812. 2015. pp. 981210–1.
• [22] Ghosh P, Bhattacharjee D, Nasipuri M. A hybrid approach to diagnosis of tuberculosis from sputum. IEEE International Conference on Electrical, Electronics, and Optimization Techniques (ICEEOT); 2016.
• [23] Sergey I, Szegedy C. Batch normalization: accelerating deep network training by reducing internal covariate shift. ICML'15 Proceedings of the 32nd International Conference on Machine Learning – vol. 37; 2015. p. 448–56.
• [24] Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Proceedings of the 25th International Conference on Neural Information Processing Systems – Vol. 1; 2012. p. 1097–105.
• [25] Glorot X, Bengio Y. Understanding the difficulty of training deep feed forward neural networks. Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics (AISTATS). Vol. 9; 2010. p. 249–56.
• [26] LeCun Y, Bottou L, Orr G, Muller K. Efficient backprop. Neural Networks: Tricks of the trade. LNCS 1524. Springer; 1998, ISBN 978-3-540-65311-0.
• [27] Costa MG, Costa Filho CFF, Junior KA, Levy PC, Xavier CM, Fujimoto LB. A sputum smear microscopy image database for automatic bacilli detection in conventional microscopy. Proc. of 36th Annual International Conference of IEEE Engineering in Medicine and Biology Society (EMBC). 2014. pp. 2841–4.
• [28] López YP, Costa Filho CFF, Aguilera LMR, Costa MGF. Automatic classification of light field smear microscopy patches using Convolutional Neural Networks for identifying Mycobacterium Tuberculosis. Proceedings of 2017 CHILEAN Conference on Electrical, Electronics Engineering, Information and Communication Technologies (CHILECON); 2017.
• [29] Froment J. Parameter-free fast pixel wise non-local means denoising. Image Process On Line 2014;4:300–26.
• [30] Otsu N. A threshold selection method from gray-level histograms. Automatica 1975;11:23–7.
• [31] Goodfellow I, Bengio Y, Courville A, editors. Deep learning. London: The MIT Press; 2016.
• [32] CostaFilho CFF, Levy PC, Xavier M, Costa MGF, Fujimoto LBM, Salem J. Micobacterium tuberculosis recognition with conventional microscopy. Proceedings of 34th Annual International IEEE Engineering in Medicine and Biology Society Conference; 2012. p. 6263–8.

Uwagi

PL

Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2018)