Fatih Ahmet ŞENEL, Veysel Atilla AYYILDIZ, Ahmet DURSUN, Kenan ÖZTÜRK

Determination of bone age using deep convolutionalneural networks

Aim: Bone age assessment is an important measure of skeletal maturity in children with a growth development disorder. Furthermore, age estimation is an important method applied in various situations such as growth observation, immigrant registration, legal criminal justice, and body detection. In this study, we aimed to develop a computer-assisted bone age detection system.Materials and Methods: This detection is usually evaluated by a trained physician using a radiological examination of the left wrist and a reference model. However, this evaluation method was stated to cause differences brought by interobserver and intraobserver variabilities. Several automated approaches have been proposed to overcome these problems, but none of them have been proven to be generalized according to different races, age ranges, and gender. Considering today's technology, it is observed that developments in the software are already used in the field of health. In this study, bone age was determined from X-rays of the left wrist using convolutional neural networks, which are a popular subject of recent years.Results: In the study, in which a total of 150 patients' images were used, different deep learning architectures were used and the results were compared. On average, the success rate was best at 98.39% with different training-testing split rates.Conclusion: This study demonstrated that deep learning could be used to determine bone age.

PDF

___

1. Spampinato C, Palazzo S, Giordano D, et al. Deep learning for automated skeletal bone age assessment in X-ray images. Med Image Anal 2017;36:41-51.
2. Pose Lepe G, Villacrés F, Fuente-Alba CS, et al. Correlation in radiological bone age determination using the Greulich and Pyle method versus automated evaluation using BoneXpert software. Rev Chil Pediatr. 2018;89:606-11.
3. Haghnegahdar A, Pakshir H, Ghanbari I. Correlation between Skeletal Age and Metacarpal Bones and Metacarpophalangeal Joints Dimensions. J Dent 2019;20:159-64.
4. Chen C, Liu Y, Kumar M, et al. Energy consumption modelling using deep learning embedded semisupervised learning. Comput Ind Eng 2019;135:757- 65.
5. Oh J, Yun K, Maoz U, et al. Identifying depression in the National Health and Nutrition Examination Survey data using a deep learning algorithm. J Affect Disord 2019;257:623-31.
6. Shen S, Sadoughi M, Chen X, et al. A deep learning method for online capacity estimation of lithium-ion batteries. J Energy Storage 2019;25:100817.
7. Wu L, Liu Z, Bera T, et al. A deep learning model to recognize food contaminating beetle species based on elytra fragments. Comput Electron Agric 2019;166:105002.
8. Yi PH, Kim TK, Wei J, et al. Automated semantic labeling of pediatric musculoskeletal radiographs using deep learning. Pediatr Radiol 2019;49:1066-70.
9. Zhu L, Zhang C, Zhang C, et al. Forming a new small sample deep learning model to predict total organic carbon content by combining unsupervised learning with semisupervised learning. Appl Soft Comput 2019;83:1055-96.
10. Cole TJ, Rousham EK, Hawley NL, et al. Ethnic and sex differences in skeletal maturation among the Birth to Twenty cohort in South Africa. Arch Dis Child. 2015;100:138-43.
11. Khan SU, Islam N, Jan Z, et al. A novel deep learning based framework for the detection and classification of breast cancer using transfer learning. Pattern Recognit Lett 2019;125:1-6.
12. Krizhevsky A, Sutskever I, Hinton GE. ImageNet Classification with Deep Convolutional Neural Networks. In: NIPS’12: Proceedings of the 25th International Conference on Neural Information Processing Systems. 2012;1097-105.
13. Szegedy C, Liu W, Jia Y, et al. Going deeper with convolutions. In: Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition. IEEE Computer Society 2015;1-9.
14. Simonyan K, Zisserman A. Very Deep Convolutional Networks for Large-Scale Image Recognition. Comput Vis Pattern Recognit 2014;1:1-14.
15. Zheng Y, Yang C, Merkulov A. Breast cancer screening using convolutional neural network and follow-up digital mammography. In: Computational Imaging III. San Francisco 2018;4.
16. Shanthi T, Sabeenian RS. Modified Alexnet architecture for classification of diabetic retinopathy images. Comput Electr Eng 2019;76:56-64.
17. Lu S, Lu Z, Zhang YD. Pathological brain detection based on AlexNet and transfer learning. J Comput Sci 2019;30:41-7.
18. Saikia AR, Bora K, Mahanta LB, et al. Comparative assessment of CNN architectures for classification of breast FNAC images. Tissue Cell 2019;57:8-14.
19. Deepak S, Ameer PM. Brain tumor classification using deep CNN features via transfer learning. Comput Biol Med 2019;111:103345.
20. Wan S, Liang Y, Zhang Y. Deep convolutional neural networks for diabetic retinopathy detection by image classification. Comput Electr Eng 2018;72:274-82.
21. Shallu, Mehra R. Breast cancer histology images classification: Training from scratch or transfer learning? ICT Express 2018;4:247-54.
22. Toraman S, Tuncer SA, Balgetir F. Is it possible to detect cerebral dominance via EEG signals by using deep learning? Med Hypotheses 2019;131:109315.
23. Koc U, Ercan I, Ozdemir S, et al. Statistical shape analysis of hand and wrist in paediatric population on radiographs. Turk J Med Sci 2020;50:1288-97.
24. Dahlberg PS, Mosdøl A, Ding Y, et al. A systematic review of the agreement between chronological age and skeletal age based on the Greulich and Pyle atlas. Eur Radiol 2019;29:2936-48.
25. Mansourvar M, Ismail MA, Raj RG, et al. The applicability of Greulich and Pyle atlas to assess skeletal age for four ethnic groups. J Forensic Leg Med 2014;22:26-9.
26. Mora S, Boechat MI, Pietka E, et al. Skeletal Age determinations in children of European and African descent: applicability of the Greulich and Pyle standards. Pediatr Res 2001;50:624-8.
27. Zhang A, Sayre JW, Vachon L, Liu BJ, Huang HK. Racial differences in growth patterns of children assessed on the basis of bone age. Radiology 2009;250:228-35
28. Fleshman K. Bone age determination in a paediatric population as an indicator of nutritional status. Trop Doct 2000;30:16-8
29. Chaumoitre K, Lamtali S, Baali A, et al. Influence of socioeconomic status and body mass index on bone age. Horm Res Paediatr 2010;74:129-35
30. Larson DB, Chen MC, Lungren MP, et al. Performance of a deep-learning neural network model in assessing skeletal maturity on pediatric hand radiographs. Radiology 2018;287:313-22.
31. Chu M, Liu B, Zhou F, et al. Bone Age Assessment Based on Two-Stage Deep Neural Networks. Int Conf Digit Image Comput Tech Appl DICTA 2018;1-6.
32. Lee H, Tajmir S, Lee J, et al. Fully Automated Deep Learning System for Bone Age Assessment. J Digit Imaging 2017;30:427-41.
33. Iglovikov VI, Rakhlin A, Kalinin AA, et al. Paediatric bone age assessment using deep convolutional neural networks. Lect Notes Comput Sci 2018;11045.
34. Schneider MT, Zhang J, Crisco JJ, et al. Men and women have similarly shaped carpometacarpal joint bones. J Biomechanics 2015;48:3420-6.
35. Didi ALM, Azman RR, Nazri M. Sex determination from carpal bone volumes: A Multi Detector Computed Tomography (MDCT) study in a Malaysian population. Legal Med 2016;20:49-52.
36. Mora S, Boechat MI, Pietka E, et al. Skeletal age determinations in children of European and African descent: applicability of the Greulich and Pyle standards. Pediatric Research 2001;50:624-8.
37. Yeniceri IO, Tosun K. Can relative area of medial clavicular epiphysis be used for age estimation? J For Med. 2017;31:53-8.