Convolutional Neural Network Architecture Densenet121 to Identify Tuberculosis

Fajri Nugraha; Sumijan S; Rini Sovia

doi:10.35134/komtekinfo.v12i4.662

Authors

Fajri Nugraha Universitas Islam Negeri Imam Bonjol Padang
Sumijan S Universitas Putra Indonesia YPTK Padang
Rini Sovia Universitas Putra Indonesia YPTK Padang

DOI:

https://doi.org/10.35134/komtekinfo.v12i4.662

Keywords:

Cigarrette Smoke, TB, Deep Learning, CNN, DenseNet121

Abstract

Smoking habits and the normalization of smoking activities are often a problem in many developing countries in the world. Cigarette smoke can cause many health problems that increase the risk of developing diseases and worsen the condition of people with the disease, one of which is Tuberculosis (TB). In Indonesia, based on the WHO Global TB Report 2024, Indonesia ranks second in the world in TB cases, it is estimated that there are more than 1,000,000 new cases every year, this disease is a very serious health problem and has obstacles in the identification process. This research aims to develop a TB disease identification system using Deep Learning. The methods used in this study are Convolutional Neural Network (CNN) and Densenet121 architecture. Convolutional Neural Network (CNN) was chosen for its ability to perform X-ray image analysis for visual validation, while Densenet121 was chosen because of its flexible architecture that can be applied to a wide range of computer vision applications, including image classification, object identification, and semantic segmentation. The research stage includes data collection, then preprocessing the image, namely resize, normalization, and conversion to arrays, then building a Convolutional Neural Network model with the selected architecture, then model training, model performance evaluation using accuracy and AUC metrics and ending with testing and validation by experts. The dataset used in this study is X-Ray data of tuberculosis patients taken from Kaggle to build a Deep Learning model that is able to identify TB through 100 chest X-ray image datasets. The results of the study show that the CNN model is able to identify tuberculosis with an accuracy rate of up to 90%, so it can help speed up early diagnosis or screening so that patients can continue to receive treatment and treatment. Therefore, the application of deep learning with the Convolutional Neural Network (CNN) method and DenseNet121 architecture based on X-Ray image data is an effective approach in the early detection of tuberculosis and seeks to make an important contribution to the control of lung diseases related to exposure to cigarette smoke in Indonesia.

References

[1] World Health Organization, "Global Tuberculosis Report 2023," Geneva: WHO, 2023. [Online]. Available: https://www.who.int/publications/i/item/9789240078306

Ministry of Health of the Republic of Indonesia, "National Report of Riskesdas 2018," Health Research and Development Agency, Jakarta, 2019. [Online].

Stop TB Partnership, "The Global Plan to End TB 2023–2030," Geneva, 2023. [Online].

C. C. Boehme et al., "Feasibility, diagnostic accuracy, and effectiveness of decentralised use of the Xpert MTB/RIF test for diagnosis of tuberculosis and multidrug resistance: a multicentre implementation study," Lancet, vol. 377, no. 9776, pp. 1495–1505, 2011. [Online]. Available: https://doi.org/10.1016/S0140-6736(11)60438-8

J. Melendez et al., "An automated tuberculosis screening strategy combining X-ray-based computer-aided detection and clinical information," Sci. Rep., vol. 6, p. 25265, 2016. [Online]. Available: https://doi.org/10.1038/srep25265

M. Lakhani and B. Sundaram, "Deep learning at chest radiography: Automated classification of pulmonary tuberculosis by using convolutional neural networks," Radiology, vol. 284, no. 2, pp. 574–582, 2017. [Online]. Available: https://doi.org/10.1148/radiol.2017162326

A. Musha et al., "A deep learning approach for COVID-19 and pneumonia detection from chest X-ray images," Int. J. Electr. Comput. Eng., vol. 12, no. 4, pp. 3655–3664, Aug. 2022. [Online]. Available: https://doi.org/10.11591/ijece.v12i4.pp3655-3664

Z. Iklima, T. M. Kadarina, and E. Ihsanto, "Realistic image synthesis of COVID-19 chest X-rays using depthwise boundary equilibrium generative adversarial networks," Int. J. Electr. Comput. Eng., vol. 12, no. 5, pp. 5444–5454, Oct. 2022. [Online]. Available: https://doi.org/10.11591/ijece.v12i5.pp5444-5454

H. M. K. Alshazly et al., "Explainable COVID-19 detection using chest CT scans and deep learning," Comput. Biol. Med., vol. 137, p. 104865, Apr. 2021. [Online]. Available: https://doi.org/10.1016/j.compbiomed.2021.104865

G. Huang, Z. Liu, L. Van Der Maaten, and K. Q. Weinberger, "Densely connected convolutional networks," in Proc. IEEE Conf. Comput. Vis. Pattern Recognit. (CVPR), 2017, pp. 4700–4708. [Online]. Available: https://doi.org/10.1109/CVPR.2017.243

S. Rajaraman et al., "Pre-trained convolutional neural networks as feature extractors toward improved TB detection in chest radiographs," IEEE Access, vol. 8, pp. 135–150, 2020. [Online]. Available: https://doi.org/10.1109/ACCESS.2020.2966885

J. Zhang et al., "COVID-19 screening on chest X-ray images using deep learning-based anomaly detection," arXiv preprint

M. Rahimzadeh and A. Attar, "A modified deep convolutional neural network for detecting COVID-19 and pneumonia from chest X-ray images," Informatics Med. Unlocked, vol. 19, p. 100360, 2020. [Online]. Available: https://doi.org/10.1016/j.imu.2020.100360

A. Abbas, M. M. Abdelsamea, and M. M. Gaber, "Classification of COVID-19 in chest X-ray images using DeTraC deep convolutional neural network," Appl. Intell., vol. 51, pp. 854–864, 2021. [Online]. Available: https://doi.org/10.1007/s10489-020-01829-7

T. H. Pham et al., "Detection of tuberculosis from chest X-ray images using deep learning," IEEE Access, vol. 8, pp. 114889–114899, 2020. [Online]. Available: https://doi.org/10.1109/ACCESS.2020.3004420

D. Pasa et al., "Efficient deep network architectures for fast chest X-ray tuberculosis screening and visualization," Sci. Rep., vol. 9, no. 1, p. 6268, 2019. [Online]. Available: https://doi.org/10.1038/s41598-019-42557-4

Y. Oh et al., "Deep learning COVID-19 features on CXR using limited training data sets," IEEE Trans. Med. Imaging, vol. 39, no. 8, pp. 2688–2700, 2020. [Online]. Available: https://doi.org/10.1109/TMI.2020.2993291

R. K. Jain et al., "Automated detection of tuberculosis using deep learning algorithms on chest X-rays," J. Med. Syst., vol. 44, no. 7, p. 127, 2020. [Online]. Available: https://doi.org/10.1007/s10916-020-01591-3

S. Jaeger et al., "Two public chest X-ray datasets for computer-aided screening of pulmonary diseases," Quant. Imaging Med. Surg., vol. 4, no. 6, pp. 475–477, 2014. [Online]. Available: https://doi.org/10.3978/j.issn.2223-4292.2014.11.20

M. D. Silva et al., "Transfer learning with deep convolutional neural networks for tuberculosis detection," in Proc. IEEE Int. Conf. Bioinformatics Biomed. (BIBM), 2018, pp. 2303–2310. [Online]. Available: https://doi.org/10.1109/BIBM.2018.8621450