Doctor Bone: training neural networks to assist in determining bone age
DOI:
https://doi.org/10.59681/2175-4411.v16.iEspecial.2024.1382Keywords:
Bone Age, Diagnostic Aid, Artificial intelligenceAbstract
Objective: To explore the application of artificial intelligence (AI) in predicting bone age from X-ray images. Method: The Interdisciplinary Methodology for the Development of Health Technologies (MIDTS) was used to develop a prediction tool. Training was conducted with convolutional neural networks (CNNs) using a dataset of 14,036 X-ray images. Results: The tool achieved a coefficient of determination (R²) of 0.94807 and a Mean Absolute Error (MAE) of 6.97, highlighting its accuracy and clinical potential. Conclusion: The project demonstrated great potential to enhance bone age prediction, with possibilities for evolution as the database grows and AI becomes more sophisticated.
References
Burns DAR, SIlva LR, Júnior DC, Blank D, Vaz EDS, Borges WG. Tratado de pediatria. 4th ed. São Paulo: Manole Saúde; 2017.
Somkantha K, Theera-Umpon N, Auephanwiriyakul S, Williamson TH. Bone age assessment in young children using automatic carpal bone feature extraction and support vector regression. J Digit Imaging. 2011 Dec;24(6):1044-58. doi: 10.1007/s10278-011-9372-3. DOI: https://doi.org/10.1007/s10278-011-9372-3
Larson DB, Chen MC, Lungren MP, Halabi SS, Stence NV, Langlotz CP. Performance of a deep-learning neural network model in assessing skeletal maturity on pediatric hand radiographs. Radiology. 2018 Apr;287(1):313-22. doi: 10.1148/radiol.2017170236. DOI: https://doi.org/10.1148/radiol.2017170236
Rijn RR, Lequin MH, Thodberg HH. Bone age assessment: automated techniques coming of age? Acta Radiol. 2013 Nov;54(9):1024-9. doi: 10.1258/ar.2012.120443. DOI: https://doi.org/10.1258/ar.2012.120443
Savi FM, de Oliveira PT, Cestari TM, Granjeiro JM, Taga R. Histomorphometric evaluation of critical-sized bone defects using osteomeasure and aperio image analysis systems. Tissue Eng Part C Methods. 2019 Dec;25(12):732-741. doi: 10.1089/ten.tec.2019.0179. DOI: https://doi.org/10.1089/ten.tec.2019.0179
Lee BD, Lee SA, Kim H, Cho J, Kim MS, Ko HK, et al. Automated bone age assessment using artificial intelligence: the future of bone age assessment. Korean J Radiol. 2021 May;22(5):792-800. doi: 10.3348/kjr.2020.0941. DOI: https://doi.org/10.3348/kjr.2020.0941
Albanese A, Stanhope R, Fitzgerald F, Preece M. The use of a computerized method of bone age assessment in clinical practice. Horm Res. 1995;44(3):2-7. doi: 10.1159/000184665. DOI: https://doi.org/10.1159/000184665
Maratová K, Chaloupková P, Šnajderová M, Krejčí H, Černá J, Novotná D, et al. A comprehensive validation study of the latest version of Bonexpert on a large cohort of Caucasian children and adolescents. Front Endocrinol (Lausanne). 2023;14:1130580. doi: 10.3389/fendo.2023.1130580. DOI: https://doi.org/10.3389/fendo.2023.1130580
Ravi D, Wong C, Deligianni F, Berthelot M, Andreu-Perez J, Lo B, Yang GZ. Deep learning for health informatics. IEEE J Biomed Health Inform. 2017 Jan;21(1):4-21. DOI: https://doi.org/10.1109/JBHI.2016.2636665
Filho JEV, Brasil CCP, Carneiro MC, Junior GS. MIDTS: método interdisciplinar para o desenvolvimento de tecnologias em saúde. In: Jorge MSB, Vergara CAC, Sampaio HADC, Moreira TMM, editors. Tecnologias e-Health em Gestão em Saúde. Curitiba: Editora CRV; 2021. p. 49-66.
Nielsen J. Ten usability heuristics [Internet]. [place unknown: publisher unknown]; 1994 Apr 24 [updated 2024 Jan 30; cited 2024 May 29]. Available from: https://www.nngroup.com/articles/ten-usability-heuristics/.
Material Design [Internet]. [place unknown: publisher unknown]; [date unknown] [cited 2024 May 29]. Available from: https://m3.material.io/.
Keras [Internet]. Keras 3 API Documentation. [place unknown: publisher unknown]; 2024 [cited 2024 May 24]. Available from: https://keras.io/api/.
Halabi SS, Prevedello LM, Kalpathy-Cramer J, Mamonov AB, Bilbily A, Cicero M, et al. The RSNA Pediatric Bone Age Machine Learning Challenge. Radiology. 2019 Feb;290(2):498-503. DOI: https://doi.org/10.1148/radiol.2018180736
ImageNet [Internet]. [place unknown: publisher unknown]; [date unknown] [cited 2024 May 28]. Available from: https://www.image-net.org/.
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Submission of a paper to Journal of Health Informatics is understood to imply that it is not being considered for publication elsewhere and that the author(s) permission to publish his/her (their) article(s) in this Journal implies the exclusive authorization of the publishers to deal with all issues concerning the copyright therein. Upon the submission of an article, authors will be asked to sign a Copyright Notice. Acceptance of the agreement will ensure the widest possible dissemination of information. An e-mail will be sent to the corresponding author confirming receipt of the manuscript and acceptance of the agreement.
