A.K. Alzubaidi1, A.A. Petrov2

1,2 Bunin Yelets State University (Yelets, Russia)

1 azhrstar90@gmail.com, 2 xeal91@yandex.ru

Diagnosis of socially significant diseases using modern intelligent neural network technologies of machine vision and classification is a promising opportunity to increase the quality of medical services while reducing cost and time costs. Of particular importance is the diagnosis of diseases such as tuberculosis. The paper provides a review analysis of existing machine learning models for classifying chest X-ray images. A transfer learning method based on the EfficientNetB0 model has been developed for the implementation of an X-ray image classifier in the diagnosis of tuberculosis. X-ray images from the publicly available X-Ray dataset are used as a test dataset. The constructed model is implemented as a Python program. A comparative analysis of the effectiveness of the implemented classifier with similar works is carried out. The aim of the work is to develop an intelligent classifier based on the EfficientNetB0 machine learning model for analyzing X-ray images in the diagnosis of tuberculosis. Obtained results can be used in the tasks of automated diagnosis of medical diseases. The results of the work are aimed at the implementation of diagnostic modules within the integrated medical information system.

Alzubaidi A.K., Petrov A.A. Using the capabilities of EfficientNetB0 machine learning model to analyze X-ray images in the diagnosis of tuberculosis. Neurocomputers. 2024. V. 26. № 5. Р. 41-48. DOI: https://doi.org/10.18127/j19998554-202405-04 (In Russian)

1. Ismailov B.I. Deep learning in medical diagnostics and control of the functions of the artificial respiration system. Biomedicine Radioengineering. 2023. V. 26. № 1. P. 5–17. DOI 10.18127/j15604136-202301-01.
2. World Health Organization. Global health observatory data on tuberculosis 2022. [Electronic resource] – Access mode: https://www.who.int/teams/global-tuberculosis-programme/tb-reports/global-tuberculosis-report-2022, date of reference 18.07.2024.
3. Steingart K.R., Flores L.L., Hopewell P.C., Dendukuri N., Schiller I., Pai M., Laal S., Ramsay A. Commercial Serological tests for the diagnosis of active pulmonary and extrapulmonary tuberculosis: An updated systematic review and Meta-Analysis. PLoS Medicine. 2011. V. 8. № 8. P. e1001062. DOI 10.1371/journal.pmed.1001062.
4. Averyanikhin A.E., Vlasov A.I., Evdokimova E.V. Hierarchical pyramidal subsampling in deep convolutional networks for visual pattern recognition. Neurocomputers. 2021. V. 23. № 1. P. 17–31. DOI 10.18127/j19998554-202101-03. (in Russian)
5. Zadorozhnaya E.A., Moroz K.A. The use of neural networks in medicine. Young researcher of the Don. 2024. V. 9. № 2. P.10–12. (in Russian)
6. Lakhani P., Sundaram B. Deep learning at chest radiography: Automated classification of pulmonary tuberculosis by using convolutional neural networks. Radiology. 2017. V. 284. № 2. P. 574–582. DOI: 10.1148/radiol.2017162326.
7. Tan M., Le Q. EfficientNet: Rethinking model scaling for convolutional neural networks. International Conference on Machine Learning. 2019. P. 6105–6114.
8. Sergeev Yu.A., Sterleva E.A., Niazyan D.A. Application of neural networks in medicine. Comparison of methods of neural network and group analysis of pathologies. StudNet. 2021. V. 4. № 9. P. 1–9. (in Russian)
9. Rajpurkar P., Irvin J., Zhu K., Yang B., Mehta H., Duan T., Ding D., Bagul A., Langlotz C., Shpanskaya K., Lungren M.P., Ng A.Y. ChexNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning. [Electronic resource] – Access mode: https://arxiv.org/pdf/1711.05225, date of reference 18.07.2024.
10. Scholle F. Deep learning in Python. St. Petersburg: Peter. 2018. 400 p. (in Russian)
11. Shen L., Margolies L.R., Rothstein J.H., Fluder E., McBride R., Sieh W. Deep learning to improve breast cancer detection on screening mammography. Scientific Reports. 2019. V. 9. №1. P. 12495.
12. Buda M., Maki A., Mazurowski M.A. A systematic study of the class imbalance problem in convolutional neural networks. Neural Networks. 2018. V. 106. P. 249–259.
13. Cross-validation: evaluating estimator performance. [Electronic resource] – Access mode: https://scikit-learn.org/dev/modules/cross_val­idation.html, date of reference 18.07.2024.
14. Abraham B., Mohan J., John S. M., Ramachandran S. Computer-aided detection of tuberculosis from X-ray images using CNN and PatternNet classifier. Journal of X-Ray Science and Technology. 2023. V. 31. № 4. P. 699–711. DOI 10.3233/xst-230028.
15. Hello A.T, Elaziz M.A., Jamal A.T., Dama R., Hassan O.F. A novel method for detection of tuberculosis in chest radiographs using artificial ecosystem-based optimization of deep neural network features. Symmetry. 2020. V. 12. № 7. P. 1146.
16. Xie Y., Wu Z., Han X., Wang H., Wu Y., Cui L., Feng J., Zhu Z., Chen Z. Computer-Aided System for the Detection of Multicategory Pulmonary Tuberculosis in Radiographs. Journal of Healthcare Engineering. 2020. P. 9205082. DOI 10.1155/2020/9205082.
17. Santosh K.C., Antani S. Automated Chest X-Ray Screening: Can Lung Region Symmetry Help Detect Pulmonary Abnormalities? IEEE Transactions on Medical Imaging. 2018. V. 37. № 5. P. 1168–1177. DOI 10.1109/TMI.2017.2775636.