Document Type : Other
Authors
- Simon Kawuma 1
- Elias Kumbakumba 2
- Vicent Mabirizi 3
- Deborah Nanjebe 4
- Kenneth Mworozi 5
- Adolf Oyesigye Mukama 6
- Lydia Kyasimire 2
1 Software Engineering, Mbarara University of Science and Technology, Mbarara, Uganda.
2 Department of Paediatrics, Mbarara University of Science and Technology, Mbarara, Uganda.
3 Department of Information Technology, Kabale University, Kabale, Uganda
4 Epicentre Mbarara Research Centre, Mbarara, Uganda.
5 Department of radiology, Mbarara University of Science and Technology, Mbarara, Uganda.
6 Medical Laboratory Science, Epicentre Mbarara Research Centre, Mbarara, Uganda.
Abstract
Tuberculosis (TB) is an underestimated cause of death in children, with only 45% of cases correctly diagnosed and reported. It is estimated that 1.12 million TB cases occurred among newborns, children, and adolescents aged less or equal 14 years. In Uganda, TB prevalence is 8.5% in children and 16.7% in adolescents. Treatment and diagnosing TB is difficulty and its high mortality rate is due to many gaps in the diagnosis of this illness especially among children. As a strategy to curb TB mortality rate in children, there exist a need to improve and expedite the screening for TB among children. Chest X-ray (CXR) are commonly used in TB burden countries like Uganda to diagnose TB patients but interpretation of the patients’ radiograph needs skilled radiologists who are few. To this end, this research aims to close the TB mortality gap in children by applying AI, primarily deep learning techniques, to detect TB in children. The study created five models, one from scratch and four transfer learning and were trained and verified using digital CXR radiograph images of children who visit the TB clinic at Mbarara Regional Referral Hospital. The model classifies clinical images of patients into normal or Tuberculosis. Transfer learning models; VGG16, VGG19, Inception V3, and ResNet50 outperformed scratch model with validation accuracy of 79.91%, 69.21%, 53.0%, 51.09% and 50.01% respectively. We hope that once the deep learning models are implemented and adopted by the radiologist, it will reduce the time spent by radiologist while analyzing CXR images.
Keywords
Main Subjects
[1] Chakaya et al., "Global Tuberculosis Report 2020– Reflections on the Global TB burden, treatment and prevention efforts," vol. 113, pp. S7-S12, 2021.
[2] M. Mandalakas et al., "Tuberculosis among children and adolescents at HIV treatment centers in sub-Saharan Africa," vol. 26, no. 12, p. 2933, 2020.
[3] P. J. Dodd, C. M. Yuen, C. Sismanidis, J. A. Seddon, and
- E. J. T. L. G. H. Jenkins, "The global burden of tuberculosis mortality in children: a mathematical modelling study," vol. 5, no. 9, pp. e898-e906, 2017.
[4] W. Tumuhimbise, A. J. G. I. R. Musiimenta, and Applications, "Barriers and Motivators of Private Hospitals’ Engagement in Tuberculosis Care in Uganda," vol. 1, no. 4, pp. 279-290, 2021.
[5] C. Marquez et al., "The age-specific burden and household and school-based predictors of child and adolescent tuberculosis infection in rural Uganda," vol. 15, no. 1, p. e0228102, 2020.
[6] U. J. N. Y. U. N. Assembly, "Political declaration of the High-Level Meeting of the General Assembly on the Fight Against Tuberculosis: resolution/adopted by the General Assembly," 2018.
[7] S. Sathitratanacheewin, P. Sunanta, and K. J. H. Pongpirul, "Deep learning for automated classification of tuberculosis- related chest X-Ray: dataset distribution shift limits diagnostic performance generalizability," vol. 6, no. 8, p. e04614, 2020.
[8] A. H. van’t Hoog et al., "Screening strategies for tuberculosis prevalence surveys: the value of chest radiography and symptoms," vol. 7, no. 7, p. e38691, 2012.
[9] E. Showkatian, M. Salehi, H. Ghaffari, R. Reiazi, and N.
- P. J. o. R. Sadighi, "Deep learning-based automatic detection of tuberculosis disease in chest X-ray images," vol. 87, no. 1, pp. 118-124, 2022.
[10] W. H. Organization, Global tuberculosis report 2013. World Health Organization, 2013.
[11] S. Modi et al., "Performance of clinical screening algorithms for tuberculosis intensified case finding among people living with HIV in Western Kenya," vol. 11, no. 12, p. e0167685, 2016.
[12] P. Rajpurkar et al., "CheXaid: deep learning assistance for physician diagnosis of tuberculosis using chest x-rays in patients with HIV," vol. 3, no. 1, pp. 1-8, 2020.
[13] F. A. Khan et al., "Computer-aided reading of tuberculosis chest radiography: moving the research agenda forward to inform policy," vol. 50, ed: Eur Respiratory Soc, 2017.
[14] Z. Z. Qin et al., "A new resource on artificial intelligence powered computer automated detection software products for tuberculosis programmes and implementers," vol. 127, p. 102049, 2021.
[15] T. Pande, M. Pai, F. A. Khan, and C. M. J. E. R. J. Denkinger, "Use of chest radiography in the 22 highest tuberculosis burden countries," vol. 46, no. 6, pp. 1816-1819, 2015.
[16] M. K. Puttagunta and S. Ravi, "Detection of Tuberculosis based on Deep Learning based methods," in Journal of Physics: Conference Series, 2021, vol. 1767, no. 1, p. 012004: IOP Publishing.
[17] R. Hooda, A. Mittal, and S. J. C. M. I. Sofat, "Tuberculosis detection from chest radiographs: a comprehensive survey on computer-aided diagnosis techniques," vol. 14, no. 4, pp. 506-520, 2018.
[18] C. Qin, D. Yao, Y. Shi, and Z. J. B. e. o. Song, "Computer-aided detection in chest radiography based on artificial intelligence: a survey," vol. 17, no. 1, pp. 1-23, 2018.
[19] U. Raghavendra, U. R. Acharya, and H. J. E. n. Adeli, "Artificial intelligence techniques for automated diagnosis of neurological disorders," vol. 82, no. 1-3, pp. 41-64, 2019.
[20] M. Adel et al., "Alzheimer’s disease computer-aided diagnosis on positron emission tomography brain images using image processing techniques," in Computer Methods and Programs in Biomedical Signal and Image Processing: IntechOpen, 2019, p. 13.
[21] M. Firmino, G. Angelo, H. Morais, M. R. Dantas, and R.
- B. e. o. Valentim, "Computer-aided detection (CADe) and diagnosis (CADx) system for lung cancer with likelihood of malignancy," vol. 15, no. 1, pp. 1-17, 2016.
[22] A. A.-A. Valliani, D. Ranti, E. K. J. N. Oermann, and therapy, "Deep learning and neurology: a systematic review," vol. 8, no. 2, pp. 351-365, 2019.
[23] I. Bakkouri, K. J. M. T. Afdel, and Applications, "Computer-aided diagnosis (CAD) system based on multi- layer feature fusion network for skin lesion recognition in dermoscopy images," vol. 79, no. 29, pp. 20483-20518, 2020.
[24] Z.-P. Jiang, Y.-Y. Liu, Z.-E. Shao, and K.-W. J. A. S. Huang, "An Improved VGG16 Model for Pneumonia Image Classification," vol. 11, no. 23, p. 11185, 2021.
[25] T. Rahman et al., "Reliable tuberculosis detection using chest X-ray with deep learning, segmentation and visualization," vol. 8, pp. 191586-191601, 2020.
[26] A. Khellal, H. Ma, and Q. Fei, "Convolutional neural network features comparison between back-propagation and extreme learning machine," in 2018 37th Chinese Control Conference (CCC), 2018, pp. 9629-9634. HIV treatment centers in sub-Saharan Africa," vol. 26, p. 2933, 2020.
[27] C. Szegedy, V. Vanhoucke, S. Ioffe, J. Shlens, and Z. Wojna, "Rethinking the inception architecture for computer vision," in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 2818-2826.
[28] I. Delibasoglu and M. J. J. o. A. R. S. Cetin, "Improved U-Nets with inception blocks for building detection," vol. 14, pp. 044512-044512, 2020.
[29] P. F. Barnes, M. A. Steele, S. M. Young, and L. A. J. C. Vachon, "Tuberculosis in patients with human immunodeficiency virus infection: how often does it mimic Pneumocystis carinii pneumonia?," vol. 102, pp. 428-432, 1992.
[30] L. M. Pinto, A. C. Shah, K. D. Shah, and Z. F. J. R. M.
- Udwadia, "Pulmonary tuberculosis masquerading as community acquired pneumonia," vol. 4, no. 3, pp. 138-140, 2011.
[31] K. Singh, S. Hyatali, S. Giddings, K. Singh, and N. J. C.
- i. c. c. Bhagwandass, "Miliary tuberculosis presenting with ARDS and shock: a case report and challenges in current management and diagnosis," vol. 2017, 2017.
[32] R. N. Rohmah, A. Susanto, I. Soesanti, and M. Tjokronagoro, "Computer Aided Diagnosis for lung tuberculosis identification based on thoracic X-ray," in 2013 International Conference on Information Technology and Electrical Engineering (ICITEE), 2013, pp. 73-78.
[33] C. Poornimadevi and H. Sulochana, "Automatic detection of pulmonary tuberculosis using image processing techniques," in 2016 International Conference on Wireless Communications, Signal Processing and Networking (WiSPNET), 2016, pp. 798-802.
[34] K. Kuppala, S. Banda, T. R. J. I. J. o. I. Barige, and D. Fusion, "An overview of deep learning methods for image registration with focus on feature-based approaches," vol. 11, pp. 113-135, 2020.
[35] S. Fatima, S. I. A. Shah, M. Z. J. I. J. o. C. Samad, and
- Engineering, "Automated tuberculosis detection and analysis using CXR’s images," vol. 10, pp. 284-290, 2018.
[36] N. Parveen, M. M. J. J. o. X.-r. S. Sathik, and Technology, "Detection of pneumonia in chest X-ray images," vol. 19, pp. 423-428, 2011.
[37] J. Xu, T. Zhao, G. Feng, M. Ni, and S. J. I. J. o. F. S. Ou, "A fuzzy C-means clustering algorithm based on spatial context model for image segmentation," vol. 23, pp. 816-832, 2021.
[38] S. J. E. S. w. A. Askari, "Fuzzy C-Means clustering algorithm for data with unequal cluster sizes and contaminated with noise and outliers: Review and development," vol. 165, p. 113856, 2021.
[39] A. Sharma, D. Raju, and S. Ranjan, "Detection of pneumonia clouds in chest X-ray using image processing approach," in 2017 Nirma University International Conference on Engineering (NUiCONE), 2017, pp. 1-4.
[40] M. Vicent, K. Simon, and S. J. I. I. P. Yonasi, "An algorithm to detect overlapping red blood cells for sickle cell disease diagnosis," vol. 16, pp. 1669-1677, 2022.
[41] E. Mohamed, K. Sirlantzis, and G. J. D. Howells, "A review of visualisation-as-explanation techniques for convolutional neural networks and their evaluation," vol. 73, p. 102239, 2022.
[42] F. Hohman, H. Park, C. Robinson, D. H. P. J. I. t. o. v. Chau, and c. graphics, "S ummit: Scaling deep learning interpretability by visualizing activation and attribution summarizations," vol. 26, pp. 1096-1106, 2019.
[43] F. Pasa, V. Golkov, F. Pfeiffer, D. Cremers, and D. J. S.
- Pfeiffer, "Efficient deep network architectures for fast chest X-ray tuberculosis screening and visualization," vol. 9, p. 6268, 2019.
[44] K. Simon, M. Vicent, K. Addah, D. Bamutura, B. Atwiine, D. Nanjebe, and A. O. Mukama, "Comparison of Deep Learning Techniques in Detection of Sickle Cell," 2023.
[45] Q. H. Nguyen, B. P. Nguyen, S. D. Dao, B. Unnikrishnan, R. Dhingra, S. R. Ravichandran, S. Satpathy, P. N. Raja, and M. C. Chua, "Deep learning models for tuberculosis detection from chest X-ray images," in 2019 26th international conference on telecommunications (ICT), 2019, pp. 381-385.
[46] L. G. C. Evalgelista and E. B. Guedes, "Computer-aided tuberculosis detection from chest X-ray images with convolutional neural networks," in Anais do XV Encontro Nacional de Inteligência Artificial e Computacional, 2018, pp. 518-527.
[47] R. Hooda, S. Sofat, S. Kaur, A. Mittal, and F. Meriaudeau, "Deep-learning: A potential method for tuberculosis detection using chest radiography," in 2017 IEEE international conference on signal and image processing applications (ICSIPA), 2017, pp. 497-502
)