Bone Age Prediction with AI Models

© 2023 by IJCTT Journal
Volume-71 Issue-2
Year of Publication : 2023
Authors : Chi-Chang Chen, Yu-Xian Chou
DOI :  10.14445/22312803/IJCTT-V71I2P104

How to Cite?

Chi-Chang Chen, Yu-Xian Chou, "Bone Age Prediction with AI Models," International Journal of Computer Trends and Technology, vol. 71, no. 2, pp. 19-24, 2023. Crossref,

Artificial intelligence (AI) models have been developed to assist in the process of bone age prediction by automating the assessment of radiographic images. These models use machine learning algorithms to learn from a dataset of previously assessed images and can then make predictions about the bone age of new images with high accuracy. In this paper, we use four AI models, namely, VGG16, ResNet50, ResNet152, and Xception, to automatically predict the bone ages of X-ray images from the Radiological Society of North America (RSNA). According to our experiments, Xception got better results than the others three models. Both the mean absolute error(mae) and median absolute error of Xception was 7.21 months. These AI models have the potential to improve the accuracy, consistency, and efficiency of bone age prediction. However, there are also limitations and challenges to using AI models for bone age prediction, such as the need for large and diverse training sets and robust validation and testing. Further research and development are needed to address these challenges and limitations to ensure that the AI models for bone age prediction are reliable and accurate in real-world settings.

Bone Age Prediction, Machine Learning, Deep Learning, Medical image processing.


[1] Xue-Lian Zhou et al., “Diagnostic Performance of Convolutional Neural Network-based Tanner-Whitehouse 3 Bone Age Assessment System,” Quantitative imaging in medicine and surgery, vol. 10, no. 3, pp. 657-667, 2020. Crossref,
[2] Byoung-Dai Lee, and Mu Sook Lee, “Automated Bone Age Assessment using Artificial Intelligence: The Future of Bone Age Assessment,” Korean Journal of Radiology, vol. 22, no. 5, pp. 792-800, 2021. Crossref,
[3] Kyu-Chong Lee et al., “Clinical Validation of a Deep Learning-based Hybrid (Greulich-Pyle and Modified Tanner-Whitehouse) Method for Bone Age Assessment,” Korean Journal of Radiology, vol. 22, no. 12, pp. 2017-2025, 2021. Crossref,
[4] RSNA Pediatric Bone Age Challenge, 2017. [Online]. Available:
[5] William Walter Greulich, and S. Idell Pyle, Radiographic Atlas of Skeletal Development of the Hand and Wrist, Stanford: Stanford University Press, 1959.
[6] Keiron O'Shea, and Ryan Nash, “An Introduction to Convolutional Neural Networks,” Neural and Evolutionary Computing, 2015. Crossref,
[7] Jiuxiang Gu et al., “Recent Advances in Convolutional Neural Networks,” Pattern Recognition, vol. 77, pp. 354-377, 2018. Crossref,
[8] Zewen Li et al., “A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects,” IEEE transactions on neural networks and learning systems, vol. 33, no. 12, pp. 6999-7019, 2022. Crossref,
[9] Safwan S. Halabi et al., “The RSNA Pediatric Bone Age Machine Learning Challenge,” Radiology, vol. 290, no. 2, pp. 498-503, 2018. Crossref,
[10] Ian Pan et al., “Improving Automated Pediatric Bone Age Estimation Using Ensembles of Models from the 2017 RSNA Machine Learning Challenge,” Radiology: Artificial Intelligence, vol. 1, no. 6, 2019. Crossref,
[11] Priscilla Whitin, and V. Jayasankar, “A Novel Deep Learning-Based System for Real-Time Temperature Monitoring of Bone Hyperthermia,” SSRG International Journal of Electrical and Electronics Engineering, vol. 10, no. 1, pp. 187-196, 2023. Crossref,
[12] Chi Fung Cheng et al., “Report of Clinical Bone Age Assessment Using Deep Learning for An Asian Population in Taiwan,” Biomedicine, vol. 11, no. 3, pp. 50-58, 2021. Crossref,
[13] Shahein H. Tajmir et al., “Artificial Intelligence-assisted Interpretation of Bone Age Radiographs Improves Accuracy and Decreases Variability,” Skeletal radiology, vol. 48, no. 2, pp. 275-283, 2019. Crossref,
[14] Luciano M. Prevedello et al., “Challenges Related to Artificial Intelligence Research in Medical Imaging and the Importance of Image Analysis Competitions,” Radiology: Artificial Intelligence, vol. 1, no. 1, 2019. Crossref,
[15] Sikender Mohsienuddin Mohammad, “AI Automation and Application in Diverse Sectors,” International Journal of Computer Trends and Technology, vol. 68, no. 1, 2020.
[16] Geert Litjens et al., “A Survey on Deep Learning in Medical Image Analysis,” Medical Image Analysis, vol. 42, pp. 60-88, 2017. Crossref,
[17] Karen Simonyan, and Andrew Zisserman, “Very Deep Convolutional Networks for Large-scale Image Recognition,” Computer Vision and Pattern Recognition, 2014. Crossref,
[18] Kaiming He et al., “Deep Residual Learning for Image Recognition,” Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770-778, 2016. Crossref,
[19] François Chollet, “Xception: Deep Learning with Depthwise Separable Convolutions,” Proceedings of The IEEE Conference on Computer Vision and Pattern Recognition, pp. 1800-1807, 2017. Crossref,
[20] J. Carreira, H. Madeira, and J.G. Silva, “Xception: A Technique for the Experimental Evaluation of Dependability in Modern Computers,” IEEE Transactions on Software Engineering, vol. 24, no. 2, pp. 125-136, 1998. Crossref,
[21] P. Nageswari, S. Rajan, and K. Manivel, “Medical Image Segmentation Approaches: A Survey,” SSRG International Journal of Electronics and Communication Engineering, vol. 7, no. 7, pp. 1-3, 2020. Crossref,
[22] Elham Beheshtian et al., “Generalizability and Bias in a Deep Learning Pediatric Bone Age Prediction Model using Hand Radiographs,” Radiology, vol. 306, no. 2, 2022. Crossref,
[23] Module: tf.keras.callbacks. [Online]. Available: