D.I. Veselov1, N.A. Andriyanov2

1,2 Financial University under the Government of the Russian Federation (Moscow, Russia)

1 diveselov@fa.ru, 2 naandriyanov@fa.ru

Cervical spine fracture requires urgent medical attention due to the fact that disruption of the integrity of one or more cervical vertebrae may lead to paralysis or death. Accurate diagnosis of spinal fracture patients using radiologic techniques is essential to initiate treatment for the patient. However, diagnosis can sometimes take a long time for physicians, so the issues of automation of the diagnostic process are of particular relevance.

The main goal of this work is to develop a computer vision system for detecting cervical spine fractures on medical images.

In the present study, a two-stage neural network algorithm was developed: in the first stage, a segmentation neural network based on UNET and the ResNet-101 convolutional neural network (CNN) was trained. In the second stage, a 3D classifier based on
ConvNext and LSTM was used. The results obtained show that the proposed algorithm is able to recognize both fractures in general and in an individual vertebra.

The developed solutions can be useful to medical workers as an auxiliary solution in the form of a decision support system, since it can process large amounts of information in a shorter time than a human.

Veselov D.I., Andriyanov N.A. Cervical spine fracture detection using artificial intelligence methods. Neurocomputers. 2024. V. 26. № 6. Р. 39-48. DOI: https://doi.org/10.18127/j19998554-202406-06 (In Russian)

1. Morozov I.N., Mlyavykh S.G. Epidemiology of spinal-spinal trauma (review). Medical Almanac. 2011. № 4(17). P. 157–159. (In Russian)
2. Andreeva T.M., Ogryzko E.V. Traumatism, orthopedic morbidity, the state of traumatological and orthopedic care for the population of Russia in 2016. Ed. acad. RAS S.P. Mironova. M.: Teler. 2017. 131 p. (In Russian)
3. Barbiellini Amidei C., Salmaso L., Bellio S., Saia M. Epidemiology of traumatic spinal cord injury: a large population-based study. Spinal Cord. 2022. V. 60. № 9. P. 812–819. DOI 10.1038/s41393-022-00795-w.
4. Tolkachev V.S., Bazhanov S.P., Ulyanov V.Yu., Fedonnikov A.S., Ninel V.G., Salihu X., Norkin I.A. The epidemiology of spine and spinal cord injuries. Saratov Scientific Medical Journal. 2018. V. 14. № 3. P. 592–595. (In Russian)
5. Sun X., He L., Jiang H., Li R., Mao W., Zhang D., Majeed Ya., Andriyanov N., Soloviev V., Fu L. Morphological estimation of primary branch length of individual apple trees during the deciduous period in modern orchard based on PointNet++. Computers and Electronics in Agriculture. 2024. V. 220. P. 108873. DOI 10.1016/j.compag.2024.108873.
6. Veselov D.I., Andriyanov N.A. Segmentation of medical images using computer vision methods. Proceedings of the XXX International Scientific and Technical Conference "Radar, navigation, communication". Voronezh: VSU Publishing House. 2024. P. 75–80. (In Russian)
7. Veselov D.I., Fan A.Ch., Gavrilin P.S., Andrianov N.A. Face detection using hybrid. Abstracts of the 21st All-Russian Conference with the International with the participation of "Mathematical methods of pattern recognition". M.: RAS. 2023. P. 146–147. (In Russian)
8. Korchagin S.A. Computer vision system for automatic classification of nanocomposites. Information-measuring and Control Systems. 2023. V. 21. № 5. P. 16−26. DOI 10.18127/j20700814-202305-03. (in Russian)
9. Galbusera F. Chapter 14 – Biomechanics of the spine. Human Orthopaedic Biomechanics. 2022. P. 265–283. DOI 10.1016/B978-0-12-824481-4.00019-6.
10. Dunsker S.B., Zhang M., Kim L., Cheong R., Cohen-Wang B., Shpanskaya K., Wetstone J., Manoj N., Rajpurkar P., Yeom K. Deep-learning artificial intelligence model for automated detection of cervical spine fracture on computed tomography (CT) imaging. Journal of Neurosurgery. 2019. V. 131. P. 12–28.
11. He K., Zhang X., Ren S., Sun J. Deep Residual Learning for Image Recognition. IEEE Conference on Computer Vision and Pattern Recognition. 2016. P. 770778. DOI 10.1109/CVPR.2016.90.
12. Chen Y., Gao Y., Li K., Zhao L., Zhao J. Vertebrae identification and localization utilizing fully convolutional networks and a hidden Markov model. IEEE Transactions on Medical Imaging. 2019. V. 39. № 2. P. 387–399. DOI 10.1109/TMI.2019.2927289.
13. Forsberg D., Sjöblom E., Sunshine J. Detection and labeling of vertebrae in MR images using deep learning with clinical annotations as training data. Journal of digital imaging. 2017. V. 30. № 4. P. 406–412. DOI: 10.1007/s10278-017-9945-x.
14. Salehinejad H., Ho E., Lin H.M., Crivellaro P., Samorodova O., Arciniegas M.T., Merali Z., Suthiphosuwan S., Bharatha A., Yeom K. Deep sequential learning for cervical spine fracture detection on computed tomography imaging. IEEE 18th International Symposium on Biomedical Imaging. 2021. P. 1911–1914. DOI 10.1109/ISBI48211.2021.9434126.
15. Chład P., Ogiela M.R. Deep learning and cloud-based computation for cervical spine fracture detection system. Electronics. 2023. V. 12. № 9. P. 2056. DOI 10.3390/electronics12092056.
16. Naguib S.M., Hamza H.M., Hosny Kh.M., Saleh M.K., Kassem M.A. Classification of Cervical Spine Fracture and Dislocation Using Refined Pre-Trained Deep Model and Saliency Map. Diagnostics. 2023. V. 13. № 7. P. 1273. DOI 10.3390/diagnostics13071273.
17. Ronneberger O., Fischer P., Brox T. U-net: Convolutional networks for biomedical image segmentation. In Medical Image Computing and Computer-Assisted Intervention. 2015. P. 234–241.
18. Kim Y.J., Ganbold B., Kim K.G. Web-based spine segmentation using deep learning in computed tomography images. Healthcare informatics research. 2020. V. 26. № 1. P. 61–67. DOI 10.4258/hir.2020.26.1.61.
19. Fang Y., Li W., Chen X., Chen K., Kang H., Yu P., Zhang R., Liao J., Hong G., Li S. Opportunistic osteoporosis screening in multi-detector CT images using deep convolutional neural networks. European Radiology. 2021. V. 31. № 4. P. 1831–1842. DOI 10.1007/ s00330-020-07312-8.
20. Lessmann N., Van Ginneken B., De Jong P., Išgum I. Iterative fully convolutional neural networks for automatic vertebra segmentation and identification. Medical image analysis. 2019. V. 53. P. 142–155. DOI 10.1016/j.media.2019.02.005.
21. Fan G., Liu H., Wu Z., Li Y., Feng C., Wang D., Luo J., Wells W.M., He S. Deep learning-based automatic segmentation of lumbosacral nerves on CT for spinal intervention: a translational study. American Journal of Neuroradiology. 2019. V. 40. № 6. P.1074–1081. DOI 10.3174/ajnr.A6070.
22. Dance D.R., Christofides S., Maidment A.D.A., McLean I.D., Ng K.H. Diagnostic radiology physics: a handbook for teachers and students. Vienna: International Atomic Energy Agency. 2014. 682 p.
23. Andriyanov N.A., Dementiev V.E., Tashlinskiy A.G. Deep Markov Models of Multidimensional Random Fields. Procedia Computer Science. 2020. V. 176. P. 1289–1298. DOI 10.1016/j.procs.2020.09.138.
24. He K., Zhang X., Ren S., Sun J. Deep Residual Learning for Image Recognition. IEEE Conference on Computer Vision and Pattern Recognition. 2016. P. 770–778. DOI 10.1109/CVPR.2016.90.
25. Todi A., Narula N., Sharma M., Gupta U. ConvNext: A Contemporary Architecture for Convolutional Neural Networks for Image Classification. 3rd International Conference on Innovative Sustainable Computational Technologies. 2023. P. 1–6. DOI 10.1109/CISCT57197. 2023.10351320.
26. Carass A., Roy S., Gherman A., Reinhold J.C., Jesson A. Evaluating White Matter Lesion Segmentations with Refined Sørensen-Dice Analysis. Scientific Reports. 2020. V. 10. № 1. P. 8242. DOI 10.1038/s41598-020-64803-w.
27. Andriyanov N.A., Dementiev V.E., Tashlinskiy A.G. Detection of objects in the images: from likelihood relationships towards scalable and efficient neural networks. Computer Optics. 2022. V. 46. № 1. P. 139–159. DOI 10.18287/2412-6179-CO-922. (In Russian)
28. Usha R., Vamsidhar Y. Binary cross entropy with deep learning technique for Image classification. International Journal of Advanced Trends in Computer Science and Engineering. 2020. V. 9. P. 121–134. DOI 10.30534/ijatcse/2020/175942020.
29. Andriyanov N.A. Andriyanov D.A. Pattern Recognition on Radar Images Using Augmentation. Ural Symposium on Biomedical Engineering, Radioelectronics and Information Technology. 2020. P. 0289–0291. DOI 10.1109/USBEREIT48449.2020.9117669.
30. Diederik K., Jimmy B. Adam: A Method for Stochastic Optimization. In Proceedings of International Conference on Learning Representations. 2014. P. 1–8.