A Cascade Model with Prior Knowledge for Bone Age Assessment
Source abstract
Abstract
Bone age is commonly used to reflect growth and development trends in children, predict adult heights, and diagnose endocrine disorders. Nevertheless, the existing automated bone age assessment (BAA) models do not consider the nonlinearity and continuity of hand bone development simultaneously. In addition, most existing BAA models are based on datasets from European and American children and may not be applicable to the developmental characteristics of Chinese children. Thus, this work proposes a cascade model that fuses prior knowledge. Specifically, a novel bone age representation is defined, which incorporates nonlinear and continuous features of skeletal development and is implemented by a cascade model. Moreover, corresponding regions of interest (RoIs) based on RUS-CHN were extracted by YOLO v5 as prior knowledge inputs to the model. In addition, based on MobileNet v2, an improved feature extractor was proposed by introducing the Convolutional Block Attention Module and increasing the receptive field to improve the accuracy of the evaluation. The experimental results show that the mean absolute error (MAE) is 4.44 months and significant correlations with the reference bone age is (r = 0.994, p < 0.01); accuracy is 94.04% for ground truth within ±1 year. Overall, the model design adequately considers hand bone development features and has high accuracy and consistency, and it also has some applicability on public datasets, showing potential for practical and clinical applications.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Human Pose and Action Recognition · Forensic Anthropology and Bioarchaeology Studies
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
College of Electrical Engineering, Sichuan University, Chengdu 610065, China
West China Second University Hospital, Sichuan University, Chengdu 610041, China
Funding
Sichuan Science and Technology Program · Chengdu Science and Technology Program · Smart Grid Sichuan Provincial Key Laboratory Emergency Key Project
References (38)
Show source-provided references
1. Artificial intelligence system can achieve comparable results to experts for bone age assessment of Chinese children with abnormal growth and development10.7717/peerj.8854 ↗
2. MABAL: A Novel Deep-Learning Architecture for Machine-Assisted Bone Age Labeling10.1007/s10278-018-0053-3 ↗
3. Normal and abnormal growth in the pediatric patient
4. Fully Automated Deep Learning System for Bone Age Assessment10.1007/s10278-017-9955-8 ↗
5. Grculich, W., and Pylc, S. (1959). Radiographic Atlas of Skeletal Development of the Hands and Wrists, Stanford University Press.
6. Automatic bone age assessment based on intelligent algorithms and comparison with TW3 method10.1016/j.compmedimag.2008.08.005 ↗
7. Nadeem, M.W., Goh, H.G., Ali, A., Hussain, M., Khan, M.A., and Ponnusamy, V.a. (2020). Bone age assessment empowered with deep learning: A survey, open research challenges and future directions. Diagnostics, 10.10.3390/diagnostics10100781 ↗
8. Hippuric acid and 3-(3-hydroxyphenyl) propionic acid inhibit murine osteoclastogenesis through RANKL-RANK independent pathway10.1002/jcp.28998 ↗
9. Tamme, R., Jürime, J., Mestu, E., Remmel, L., Purge, P., Mengel, E., and Tillmann, V. (2019). Physical Activity in Puberty Is Associated with Total Body and Femoral Neck Bone Mineral Characteristics in Males at 18 Years of Age. Medicina, 55.10.3390/medicina55050203 ↗
10. Zhang, S., and Liu, L. (2022, May 04). The Skeletal Development Standards of Hand and Wrist for Chinese Children China 05 1. TW3-C RUS, TW3-C Carpal, and RUS-CHN Methods. Chin. J. Sport. Med., Available online: https://pesquisa.bvsalud.org/portal/resource/pt/wpr-587292.
11. A comparative study of the consistency between the Greulich-Pyle method and the China 05 method in normal children aged 6–18 years
12. The BoneXpert method for automated determination of skeletal maturity10.1109/TMI.2008.926067 ↗
13. Bone Age Assessment Based on Rank-Monotonicity Enhanced Ranking CNN10.1109/ACCESS.2019.2937341 ↗
14. A deep automated skeletal bone age assessment model with heterogeneous features learning10.1007/s10916-018-1091-6 ↗
15. Computer-assisted bone age assessment: Image preprocessing and epiphyseal/metaphyseal ROI extraction10.1109/42.938240 ↗
16. Dallora, A.L., Anderberg, P., Kvist, O., Mendes, E., Diaz Ruiz, S., and Sanmartin Berglund, J. (2019). Bone age assessment with various machine learning techniques: A systematic literature review and meta-analysis. PLoS ONE, 14.10.1371/journal.pone.0220242 ↗
17. Assessing the bone age of children in an automatic manner newborn to 18 years range10.1007/s10278-019-00209-z ↗
18. Deep learning for automated skeletal bone age assessment in X-ray images10.1016/j.media.2016.10.010 ↗
19. TW3-based fully automated bone age assessment system using deep neural networks10.1109/ACCESS.2019.2903131 ↗
20. Ridge regression neural network for pediatric bone age assessment10.1007/s11042-021-10935-8 ↗
21. Regression convolutional neural network for automated pediatric bone age assessment from hand radiograph10.1109/JBHI.2018.2876916 ↗
22. Skeletal bone age prediction based on a deep residual network with spatial transformer10.1016/j.cmpb.2020.105754 ↗
23. Iglovikov, V.I., Rakhlin, A., Kalinin, A.A., and Shvets, A.A. (2018). Paediatric bone age assessment using deep convolutional neural networks. Deep Learning in mEdical Image Analysis and Multimodal Learning for Clinical Decision Support, Springer.10.1101/234120 ↗
24. Deep age estimation: From classification to ranking10.1109/TMM.2017.2786869 ↗
25. Detection of mold on the food surface using YOLOv510.1016/j.crfs.2021.10.003 ↗
26. Sandler, M., Howard, A., Zhu, M., Zhmoginov, A., and Chen, L.C. (2018, January 18–23). Mobilenetv2: Inverted residuals and linear bottlenecks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Salt Lake City, UT, USA.10.1109/CVPR.2018.00474 ↗
27. Dynamic Receptive Field-Based Object Detection in Aerial Imaging10.3788/AOS202040.0415001 ↗
28. Woo, S., Park, J., Lee, J.Y., and Kweon, I.S. (2018, January 8–24). Cbam: Convolutional block attention module. Proceedings of the European Conference on Computer Vision (ECCV), Munich, Germany.10.1007/978-3-030-01234-2_1 ↗
29. Zhang, C., Liu, S., Xu, X., and Zhu, C. (2019, January 15–20). C3AE: Exploring the limits of compact model for age estimation. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.10.1109/CVPR.2019.01287 ↗
30. Generative adversarial network based data augmentation and gender-last training strategy with application to bone age assessment10.1016/j.cmpb.2021.106456 ↗
31. Improving accuracy of lung nodule classification using deep learning with focal loss10.1155/2019/5156416 ↗
32. Mo, X., Wei, T., Zhang, H., Huang, Q., and Luo, W. (2019, January 26–29). Label-Smooth Learning for Fine-Grained Visual Categorization. Proceedings of the Asian Conference on Pattern Recognition, Auckland, New Zealand.10.1007/978-3-030-41404-7_2 ↗
33. Zulkifley, M.A., Abdani, S.R., and Zulkifley, N.H. (2020). Automated bone age assessment with image registration using hand X-ray images. Appl. Sci., 10.10.3390/app10207233 ↗
34. Miao, J., Yue, H., Wu, X., Xu, D., and Chen, W. (2020, January 6–8). Bone age assessment based on SuperPoint features. Proceedings of the 2020 Chinese Automation Congress (CAC), Shanghai, China.10.1109/CAC51589.2020.9327736 ↗
35. Radiographs and texts fusion learning based deep networks for skeletal bone age assessment10.1007/s11042-020-08943-1 ↗
36. Bone age assessment and sex determination using transfer learning10.1016/j.eswa.2022.116926 ↗
37. Hyperparameter optimization based deep convolution neural network model for automated bone age assessment and classification10.1016/j.displa.2022.102206 ↗
38. Age Assessment of Youth and Young Adults Using Magnetic Resonance Imaging of the Knee: A Deep Learning Approach10.2196/16291 ↗