Performance Evaluation of Deep Learning Methods for Cervical Spine Fracture Detection
DOI:
https://doi.org/10.22399/ijcesen.4671Keywords:
Cervical Spine Fracture, AI in Medical Image Analysis, Deep Learning, CNN, Resnet, Object DetectionAbstract
Cervical spine fractures are critical medical emergencies with the possible to cause permanent paralysis or death if not promptly identified and treated. This study addresses the gap in leveraging deep learning models for their detection by proposing a two-stage pipeline using a curated dataset of computed tomography (CT) images comprising both fractured and non-fractured cases. In the first stage, cervical vertebrae are identified within CT image slices using a multi-input network based on the Global Context Vision Transformer (GC ViT) architecture, benchmarked against leading Deep Learning models. During the second phase, multiple architectures such as Convolutional Neural Networks (CNNs) were employed to perform the designated classification tasks (CNN), ResNet, EnlightenmentNet, and a hybrid CNN+ResNet architecture, are employed to detect fractures, achieving accuracy of 64%, 70%, 54%, and 94%, respectively, with further comparisons against YOLOv5 for localization performance. The hybrid CNN+ResNet model demonstrated superior accuracy, significantly reducing radiologists' workload and enhancing diagnostic precision and efficiency. This approach has substantial clinical implications, offering a promising solution to improve patient outcomes through timely and reliable detection of cervical spine fractures
References
[1] Donald A Neumann. Kinesiology of the musculoskeletal system-e-book: foundations for rehabilitation. Elsevier Health Sciences, 2016.
[2] AF Voter, ME Larson, JW Garrett, and J-PJ Yu. Diagnostic accuracy and failure mode analysis of a deep learning algorithm for the detection of cervical spine fractures. American Journal of Neuroradiology, 42(8):1550–1556, 2021. DOI: https://doi.org/10.3174/ajnr.A7179
[3] Claudio Barbiellini Amidei, Laura Salmaso, Stefania Bellio, and Mario Saia. Epidemiology of traumatic spinal cord injury: a large population-based study. Spinal Cord, 60(9):812–819, 2022. DOI: https://doi.org/10.1038/s41393-022-00795-w
[4] Christian Zanza, Gilda Tornatore, Cristina Naturale, Yaroslava Longhitano, Angela Saviano, Andrea Piccioni, Aniello Maiese, Michela Ferrara, Gianpietro Volonnino, Giuseppe Bertozzi, et al. Cervical spine injury: Clinical and medico-legal overview. La radiologia medica, 128(1):103–112, 2023. DOI: https://doi.org/10.1007/s11547-022-01578-2
[5] Fabio Galbusera. Biomechanics of the spine. In Human Orthopaedic Biomechanics, pages 265–283. Elsevier,2022. DOI: https://doi.org/10.1016/B978-0-12-824481-4.00019-6
[6] Emmanuel Montagnon, Milena Cerny, Alexandre Cadrin-Chênevert, Vincent Hamilton, Thomas Derennes, André Ilinca, Franck Vandenbroucke-Menu, Simon Turcotte, Samuel Kadoury, and An Tang. Deep learning work flow in radiology: a primer. Insights into imaging, 11:1–15, 2020. DOI: https://doi.org/10.1186/s13244-019-0832-5
[7] Yizhi Chen, Yunhe Gao, Kang Li, Liang Zhao, and Jun Zhao. Vertebrae identifi cation and localization utilizing fully convolutional networks and a hidden markov model. IEEE Transactions on Medical Imaging, 39(2):387–399,2019. DOI: https://doi.org/10.1109/TMI.2019.2927289
[8] Daniel Forsberg, Erik Sjöblom, and Jeffrey L Sunshine. Detection and labeling of vertebrae in mr images using deep learning with clinical annotations as training data. Journal of digital imaging, 30(4):406–412, 2017. DOI: https://doi.org/10.1007/s10278-017-9945-x
[9] Hojjat Salehinejad, Edward Ho, Hui-Ming Lin, Priscila Crivellaro, Oleksandra Samorodova, Monica Tafur Arciniegas, Zamir Merali, Suradech Suthiphosuwan, Aditya Bharatha, Kristen Yeom, et al. Deep sequential learning for cervical spine fracture detection on computed tomography imaging. In 2021 IEE E 18th International Symposium on Biomedical Imaging (ISBI), pages 1911–1914. IEEE, 2021.
[10] Paweł Chł ˛ad and Marek R Ogiela. Deep learning and cloud-based computation for cervical spine fracture detection system. Electronics, 12(9):2056, 2023.
[11] Soaad M Naguib, Hanaa M Hamza, Khalid M Hosny, Mohammad K Saleh, and Mohamed A Kassem. Classifi ca- tion of cervical spine fracture and dislocation using refi ned pre-trained deep model and saliency map. Diagnostics, 13(7):1273, 2023. DOI: https://doi.org/10.3390/diagnostics13071273
[12] Showmick Guha Paul, Arpa Saha, and Md Assaduzzaman. A real-time deep learning approach for classifying cervical spine fractures. Healthcare Analytics, page 100265, 2023. DOI: https://doi.org/10.1016/j.health.2023.100265
[13] Olaf Ronneberger, Philipp Fischer, and Thomas Brox. U-net: Convolutional networks for biomedical image seg- mentation. In Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18, pages 234–241. Springer, 2015. DOI: https://doi.org/10.1007/978-3-319-24574-4_28
[14] Young Jae Kim, Bilegt Ganbold, and Kwang Gi Kim. Web-based spine segmentation using deep learning in computed tomography images. Healthcare informatics research, 26(1):61–67, 2020. DOI: https://doi.org/10.4258/hir.2020.26.1.61
[15] Yijie Fang, Wei Li, Xiaojun Chen, Keming Chen, Han Kang, Pengxin Yu, Rongguo Zhang, Jianwei Liao, Guobin Hong, and Shaolin Li. Opportunistic osteoporosis screening in multi-detector ct images using deep convolutional neural networks. European Radiology, 31:1831–1842, 2021. DOI: https://doi.org/10.1007/s00330-020-07312-8
[16] Nikolas Lessmann, Bram Van Ginneken, Pim A De Jong, and Ivana Išgum. Iterative fully convolutional neural networks for automatic vertebra segmentation and identifi cation. Medical image analysis, 53:142–155, 2019. DOI: https://doi.org/10.1016/j.media.2019.02.005
[17] Guoxin Fan, Huaqing Liu, Zhenhua Wu, Yufeng Li, Chaobo Feng, Dongdong Wang, Jie Luo, WM Wells, and Shisheng He. Deep learning–based automatic segmentation of lumbosacral nerves on ct for spinal intervention: a translational study. American Journal of Neuroradiology, 40(6):1074–1081, 2019. DOI: https://doi.org/10.3174/ajnr.A6070
[18] H. Salehinejad et al., “Deep sequential learning for cervical spine fracture detection on computed tomography imaging,” in 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI), 2021, pp. 1911–1914. DOI: https://doi.org/10.1109/ISBI48211.2021.9434126
[19] B. J. Erickson, P. Korfiatis, T. L. Kline, Z. Akkus, K. Philbrick, and A. D. Weston, “Deep learning in radiology: Does one size fit all?” J. Amer. College Radiol., vol. 15, no. 3, pp. 521–526, Mar. 2018. DOI: https://doi.org/10.1016/j.jacr.2017.12.027
[20] Chład and Ogiela “Deep Learning and Cloud-Based Computation for Cervical Spine Fracture Detection System”, 2023, 12(9), 2056. DOI: https://doi.org/10.3390/electronics12092056
[21] Z. Krawczyk and J. Starzynski, “Bones detection in the pelvic area on the basis of YOLO neural network,” in Proc. 19th Int. Conf. Comput. Problems Electr. Eng., Sep. 2018. DOI: https://doi.org/10.1109/CPEE.2018.8506970
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 International Journal of Computational and Experimental Science and Engineering
This work is licensed under a Creative Commons Attribution 4.0 International License.
