Combined algorithm for bone age determination based on hand X-rays analysis

  • Alexander M. Nedzved Belarusian State University, 4 Niezaliežnasci Avenue, Minsk 220030, Belarus; United Institute of Informatics Problems, National Academy of Sciences of Belarus, 6 Surhanava Street, Minsk 220012, Belarus https://orcid.org/0000-0001-6367-5900
  • Ivan I. Kosik Belarusian State Medical University, 83 Dziaržynskaha Avenue, Minsk 220116, Belarus
  • Ryhor M. Karapetsian Belarusian State Medical University, 83 Dziaržynskaha Avenue, Minsk 220116, Belarus

Abstract

In this paper, we investigate the urgent problem associated with bone age determination using hand X-rays. A combined algorithm for the recognition of radiographs is proposed, which uses simultaneous two neural network models, based on Xception and DenseNet169. The method allows to generalize the knowledge of different medical experts and increases the accuracy of bone age prediction in general.

Author Biographies

Alexander M. Nedzved, Belarusian State University, 4 Niezaliežnasci Avenue, Minsk 220030, Belarus; United Institute of Informatics Problems, National Academy of Sciences of Belarus, 6 Surhanava Street, Minsk 220012, Belarus

doctor of science (engineering); dean of the faculty of applied mathematics and computer science, Belarusian State Medical University, and chief researcher at the department of intellectual information systems, United Institute of Informatics Problems, National Academy of Sciences of Belarus

Ivan I. Kosik, Belarusian State Medical University, 83 Dziaržynskaha Avenue, Minsk 220116, Belarus

researcher at the laboratory of information and computer technologies of research division

Ryhor M. Karapetsian, Belarusian State Medical University, 83 Dziaržynskaha Avenue, Minsk 220116, Belarus

head of the laboratory of information and computer technologies of research division

References

  1. Baykal E, Dogan H, Ercin ME, Ersoz S, Ekinci M. Transfer learning with pre-trained deep convolutional neural networks for serous cell classification. Multimedia Tools and Applications. 2020;79(21–22):15593–15611. DOI: 10.1007/s11042-019-07821-9.
  2. Deng J, Dong W, Socher R, Li L-J, Li K, Fei-Fei L. ImageNet: a large-scale hierarchical image database. In: IEEE conference on computer vision and pattern recognition; 2009 June 20–25; Miami, Florida, USA. [S. l.]: Institute of Electrical and Electronics Engineers; 2009. p. 248–255. DOI: 10.1109/CVPR.2009.5206848.
  3. Gulli A, Pal S. Deep learning with Keras: implement neural networks with Keras on Theano and TensorFlow. Birmingham: Packt Publishing; 2017. 320 p. Russian edition: Gulli A, Pal S. Biblioteka Keras – instrument glubokogo obucheniya: realizatsiya neironnykh setei s pomoshch’ yu bibliotek Theano i TensorFlow. Slinkin AA, translator. Moscow: DMK Press; 2017. 294 p.
  4. Chollet F. Xception: deep learning with depthwise separable convolutions. In: IEEE conference on computer vision and pattern recognition (CVPR); 2017 July 21–26; Honolulu, Hawaii, USA. [S. l.]: Institute of Electrical and Electronics Engineers; 2017. p. 1800–1807. DOI: 10.1109/CVPR.2017.195.
  5. Huang G, Liu Z, Van Der Maaten L, Weinberger KQ. Densely connected convolutional networks. In: IEEE conference on computer vision and pattern recognition (CVPR); 2017 July 21–26; Honolulu, Hawaii, USA. [S. l.]: Institute of Electrical and Electronics Engineers; 2017. p. 2261–2269. DOI: 10.1109/CVPR.2017.243.
  6. Khan A, Sohail A, Zahoora U, Qureshi AS. A survey of the recent architectures of deep convolutional neural networks. Artificial Intelligence Review. 2020:1–70. DOI: 10.1007/s10462-020-09825-6.
  7. Lin T-Y, Dollár P, Girshick R, He K, Hariharan B, Belongie S. Feature pyramid networks for object detection. In: IEEE conference on computer vision and pattern recognition (CVPR); 2017 July 21–26; Honolulu, Hawaii, USA. [S. l.]: Institute of Electrical and Electronics Engineers; 2017. p. 936–944. DOI: 10.1109/CVPR.2017.106.
  8. Machine learning: concepts, methodologies, tools and applications. Hershey: Information Science Reference; 2011. 2141 p. (Premier reference source).
  9. Jordan MI, Jacobs RA. Hierarchical mixtures of experts and the EM algorithm. Neural Computation. 1994;6(2):181–214. DOI: 10.1162/neco.1994.6.2.181.
  10. Wolpert DH. Stacked generalization. Neural Networks. 1992;5(2):241–259. DOI: 10.1016/S0893-6080(05)80023-1.
  11. Menahem E, Rokach L, Elovici Y. Troika – an improved stacking schema for classification tasks. Information Sciences. 2009; 179(24):4097–4122. DOI: 10.1016/j.ins.2009.08.025.
  12. Seewald AK. How to make stacking better and faster while also taking care of an unknown weakness. In: Sammut C, Hoffmann AG, editors. ICML’02. Proceedings of the 19 th International conference on machine learning. San Francisco: Morgan Kaufmann Publishers; 2002. p. 554–561.
Keywords: bone age, X-ray, radiograph, image processing, activation map, convolutional neural network, automation  of diagnostics
Supporting Agencies This research was partially supported by the project of the Belarusian Republican Foundation for Fundamental Research and Russian Foundation for Basic Research F20R-134 «Development and research of descriptive methods for analyzing dynamic images to automatisation diagnostic procedures» and project of the State Committee on Science and Technology of the Republic of Belarus and Ministry of Science and Technology of the People’s Republic of China F20KITG-006 «Motion analysis of biological objects in video-sequence obtained by high-resolution microscope».
How to Cite
Nedzved, A. M., Kosik, I. I., & Karapetsian, R. M. (1). Combined algorithm for bone age determination based on hand X-rays analysis. Journal of the Belarusian State University. Mathematics and Informatics, 2, 105-114. https://doi.org/10.33581/2520-6508-2020-2-105-114
Section
Theoretical Foundations of Computer Science