Vol. 29 No. 2 (2026) Cover Image
Vol. 29 No. 2 (2026)

Published: June 20, 2026

Pages: 392-401

Articles

Standardizing Cervical Spine MRI for AI: A Mathematically Formalized Preprocessing and Harmonization Pipeline

Noel John Veigas 1 Dasharathraj K Shetty 1 Shyamasunder Bhat N 2 Prithvishree Ravindra 3 Praveen Shastry 1
1 Manipal Institute of Technology, Manipal Academy of Higher Education, Manipal, India
2 Department of Orthopedics, Kasturba Medical College, Manipal Academy of Higher Education, Manipal - 576104, India.
3 Department of Emergency Medicine, Kasturba Medical College, Manipal Academy of Higher Education, Manipal, India
History
  • Received: March 4, 2026
  • Revised: April 26, 2026
  • Accepted: May 2, 2026
  • Available Online: June 20, 2026
DOI

https://doi.org/10.29194/NJES.29020392

Abstract

The standardization and preprocessing decisions determine the performance, reproducibility and generalizability of Artificial Intelligence (AI) models in cervical spine Magnetic Resonance Imaging (MRI). In this research, the author suggests a six-step standardized preprocessing pipeline that can help with the reliable analysis of heterogeneous MRI data using AI. It is obtained by the pipeline based on recurrent methodological patterns found in the literature, focusing on the operations reported to stabilize image intensity distribution and reduce unwanted sources of variability. A systematic literature review was screened on 2,882 records, and 43 articles were included per the inclusion criteria. Due to the large variability of imaging procedures, preprocessing plans, and reported outcomes metrics, no quantitative meta-analysis was performed and a qualitative synthesis was done. Basic methods such as segmentation, denoising, intensity normalization, augmentation, bias-field correction were consistently reported to improve AI performance. Studies evaluating tasks such as normalization and automated segmentation reported accuracies of 94% to 99.95%. Moreover, less-specific harmonization methods like Combatting Batch Effects (ComBat), z-score normalization and histogram matching, and Generative Adversarial Network (GAN)-based ones were also often attributed with lower scanner- and site-based variations. Regardless of these improvements, the AI workflow remains unstable because of the irregularity of predefined parameters, unequal standards of acquisition, and lack of methodological reporting. In efforts to deal with such challenges, the current work proposes a methodical six-stage preprocessing model that consists of quality control, noise removal, intensity normalization, anatomical segmentation, harmonization and validation. The suggested workflow model offers a practical and transparent basis of clinically translatable AI usage in cervical spine MRI based on conventional mathematical equations and clear documentation.

Keywords

References

  1. M. Yaseen, M. Ali, S. Ali, A. Hussain, M.-I. Joo, and H.-C. Kim, “Cervical spine fracture detection and classification using Two-Stage Deep Learning methodology,” IEEE Access, vol. 12, pp. 72131–72142, (2024). https://doi.org/10.1109/access.2024.3398061
  2. Q. Zhang, X. Chen, Z. He, L. Wu, K. Wang, J. Sun, and H. Shen, “Pathology-guided AI system for accurate segmentation and diagnosis of cervical spondylosis,” IEEE Journal of Biomedical and Health Informatics, vol. 30, no. 2, pp. 1216–1229, Feb. (2026). https://doi.org/10.1109/JBHI.2025.3598469
  3. S. Niyoteka, R. Rouhi, P.-A. Laurent, S. Achkar, A. Carré, S. Diffetocq, C. Balleyguier, C. Chargari, E. Deutsch, and C. Robert, “GAN-based standardization of MR images: A promising approach for the development of multicenter radiomic models,” Physics in Medicine & Biology, vol. 70, no. 18, (2025). https://doi.org/10.1088/1361-6560/adf2f4
  4. M. L. Kumar, P. Sampath, P. V. V. S. Srinivas, D. Dommeti, and S. R. Nallapati, “Leveraging Deep Learning for Accurate Detection and Precise Localization of Vertebral Fractures in Medical Imaging,” in 2023 4th International Conference on Smart Electronics and Communication (ICOSEC), Trichy, (2023). https://doi.org/10.1109/ICOSEC58147.2023.10275972
  5. C. Kim, S.-M. Park, S. Lee, and D. Lee, “A deep learning harmonization of multi-vendor MRI for robust intervertebral disc segmentation,” IEEE Access, vol. 12, pp. 19482 - 19499, (2024). https://doi.org/10.1109/ACCESS.2024.3360272
  6. A. Lee et al., “Using deep learning to enhance reporting efficiency and accuracy in degenerative cervical spine MRI,” The Spine Journal, vol. 25, no. 9, pp. 1942–1950, (2025). https://doi.org/10.1016/j.spinee.2025.03.009
  7. M. Singh, U. Tripathi, K. K. Patel, K. Mohit, and S. Pathak, “An efficient deep learning based approach for automated identification of cervical vertebrae fracture as a clinical support aid,” Scientific Reports, vol. 15, Art. no. 25651, (2025). https://doi.org/10.1038/s41598-025-10448-6.
  8. E. A. Tarakçı et al., “Novel preprocessing-based sequence for comparative MR cervical lymph node segmentation,” Journal of Clinical Medicine, vol. 14, no. 6, p. 1802, (2025). http://doi.org/10.3390/jcm14061802
  9. J. Park, J. Yang, S. Park, and J. Kim, “Deep learning-based approaches for classifying foraminal stenosis using cervical spine radiographs,” Electronics, vol. 12, no. 1, p. 195, (2023). http://doi.org/10.3390/electronics12010195
  10. V. Viswan, F. Hajamohideen, K. Subramanian, N. Shaffi, and M. Mahmud, “Enhancing insights: Unravelling the potential of preprocessing MRI for artificial intelligence based Alzheimer’s disease classification,” in Machine Learning Models and Architectures for Biomedical Signal Processing. Academic Press, pp. 125–151, (2025). http://doi.org/10.1016/B978-0-443-22158-3.00006-5
  11. X. Feng et al., “Feasibility of fully automatic assessment of cervical canal stenosis using MRI via deep learning,” Quantitative Imaging in Medicine and Surgery, vol. 15, no. 9, pp. 8457–8470, (2025). http://doi.org/10.21037/qims-2025-67
  12. C.-W. Chen, H.-H. Tsai, C.-Y. Yeh, C.-K. Yang, H.-K. Tsou, P.-Y. Leong, and J. C.-C. Wei, “Application of artificial intelligence in rheumatic disease classification: An example of ankylosing spondylitis severity inspection model,” Annals of Medicine, vol. 57, no. 1, Art. no. 2512131, (2025). https://doi.org/10.1080/07853890.2025.2512131
  13. M. A. Larhmam, M. Benjelloun, and S. Mahmoudi, “Vertebra identification using template matching model and k-means clustering,” International Journal of Computer Assisted Radiology and Surgery, vol. 9, pp. 177–187, (2014). http://doi.org/10.1007/s11548-013-0927-2
  14. F. Muhammad et al., “Magnetization transfer MRI (MT-MRI) detects white matter damage beyond the primary site of compression in degenerative cervical myelopathy using a novel semi-automated analysis,” Computers in Biology and Medicine, vol. 197, pt. B, Art. no. 111083, (2025). http://doi.org/10.1016/j.compbiomed.2025.111083
  15. K. Nispel, T. Lerchl, G. Gruber, H. Moeller, R. Graf, V. Senner, and J. S. Kirschke, “From MRI to FEM: an automated pipeline for biomechanical simulations of vertebrae and intervertebral discs,” Frontiers in Bioengineering and Biotechnology, vol. 12, Art. no. 1485115, (2025). http://doi.org/10.3389/fbioe.2024.1485115
  16. E. Kondrateva et al., “Negligible effect of brain MRI data preprocessing for tumor segmentation,” Biomedical Signal Processing and Control, vol. 96, part A, p. 106599, Oct. (2024). http://doi.org/10.1016/j.bspc.2024.106599
  17. B. De Leener et al., “SCT: Spinal Cord Toolbox, an open-source software for processing spinal cord MRI data,” NeuroImage, vol. 145, pp. 24–43, (2017). http://doi.org/10.1016/j.neuroimage.2016.10.009
  18. S. Patel and D. Kumar, “Predictive identification of oral cancer using AI and machine learning,” Oral Oncol. Rep., vol. 13, (2025). https://doi.org/10.1016/j.oor.2024.100697
  19. N. Bhagwat et al., “Understanding the impact of preprocessing pipelines on neuroimaging cortical surface analyses,” GigaScience, vol. 10, no. 1, (2021). https://doi.org/10.1093/gigascience/giaa155
  20. K. B. De Raad et al., “The effect of preprocessing on convolutional neural networks for medical image segmentation,” 2021 IEEE 18th International Symposium on Biomedical Imaging (ISBI), (2021). https://doi.org/10.1109/ISBI48211.2021.9433952
  21. M. Juneja, N. Aggarwal, S. K. Saini, S. Pathak, M. Kaur, and M. Jaiswal, “A comprehensive review on artificial intelligence-driven preprocessing, segmentation, and classification techniques for precision furcation analysis in radiographic images,” Multimedia Tools and Applications, vol. 84, pp. 21467–21520, (2024) https://doi.org/10.1007/s11042-024-19920-3
  22. G. Ghobrial and C. Roth, “Deep learning-based automated segmentation and quantification of the dural sac cross-sectional area in lumbar spine MRI,” Frontiers in Radiology, vol. 5, Art. no. 1503625, (2025). https://doi.org/10.3389/fradi.2025.1503625
  23. X. Chang, X. Cai, Y. Dan, Y. Song, Q. Lu, G. Yang, and S. Nie, “Self-supervised learning for multi-center magnetic resonance imaging harmonization without traveling phantoms,” Physics in Medicine & Biology, vol. 67, no. 14, (2022). https://doi.org/10.1088/1361-6560/ac7b66
  24. M. Seif, T. Leutritz, S. Schading, T. Emmengger, A. Curt, N. Weiskopf, and P. Freund, “Reliability of multi-parameter mapping (MPM) in the cervical cord: A multi-center multi-vendor quantitative MRI study,” NeuroImage, vol. 264, p. 119751, (2022). https://doi.org/10.1016/j.neuroimage.2022.119751
  25. W. Y. Zhang and H. J. Chen, “Applications and prospects of artificial intelligence in cervical spine surgery,” Zhonghua Wai Ke Za Zhi, vol. 63, no. 5, pp. 383–388, (2025). http://doi.org/10.3760/cma.j.cn112139-20250104-00007
  26. T. W. Henry et al., “Is the atlantoaxial level overlooked in the radiologic interpretation of cervical magnetic resonance imaging?” Clin. Spine Surg., vol. 38, no. 10, pp. E568–E572, 2025). http://doi.org/10.1097/BSD.0000000000001805
  27. M. Pekmezci, A. A. Theologis, R. Dionisio, R. Mackersie, and R. T. McClellan, “Cervical spine clearance protocols in Level I, II, and III trauma centers in California,” Spine J., vol. 15, no. 3, pp. 398–404, (2015). http://doi.org/10.1016/j.spinee.2014.12.142
  28. P. Woznicki, F. C. Laqua, A. Al-Haj, T. Bley, and B. Baeßler, “Addressing challenges in radiomics research: systematic review and repository of open-access cancer imaging datasets,” Insights Imaging, vol. 14, no. 1, p. 216, (2023). http://doi.org/10.1186/s13244-023-01556-w
  29. J. P. Harper, G. R. Lee, I. Pan, X. V. Nguyen, N. Quails, and L. M. Prevedello, “External validation of a winning artificial intelligence algorithm from the RSNA 2022 cervical spine fracture detection challenge,” AJNR Am. J. Neuroradiol., vol. 46, no. 9, pp. 1852–1858, (2025). http://doi.org/10.3174/ajnr.A8715
  30. C. J. Entenmann, K. Kersting, P. Vajkoczy, and A. Zdunczyk, “Closing the diagnostic gap: A narrative review of recent advances in functional MRI diagnostics in spinal cord injury,” Brain Spine, vol. 5, p. 104283, (2025). http://doi.org/10.1016/j.bas.2025.104283
  31. M. V. C. Souza, R. Z. V. Costa, and K. E. P. Pinho, “Comparative analysis of parameters of magnetic resonance examinations for the accreditation process of a diagnostic image center in Curitiba/Brazil,” in XXVII Brazilian Congress on Biomedical Engineering, Springer International Publishing, (2022). [Online]. Available: https://www.springerprofessional.de/comparative-analysis-of-parameters-of-magnetic-resonance-examina/20313580 (Accessed: Mar. 23, 2026)
  32. V. A. Magnotta, L. Friedman, and F BIRN, “Measurement of signal-to-noise and contrast-to-noise in the fBIRN multicenter imaging study,” J. Digit. Imaging, vol. 19, no. 2, pp. 140–147,(2006). https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3045184
  33. P. Perona and J. Malik, “Scale-space and edge detection using anisotropic diffusion,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 12, no. 7, pp. 629–639, (1990). https://ieeexplore.ieee.org/document/56205
  34. A. Jain, K. Nandakumar, and A. Ross, “Score normalization in multimodal biometric systems,” Pattern Recogn., vol. 38, no. 12, pp. 2270–2285, (2005). https://doi.org/10.1016/j.patcog.2005.01.012
  35. O. Ronneberger, P. Fischer, and T. Brox, “U-Net: Convolutional networks for biomedical image segmentation,” Medical Image Computing and Computer-Assisted Intervention – MICCAI 2015, pp. 234–241, (2015). https://doi.org/10.1007/978-3-319-24574-4_28
  36. R. C. Gonzalez and R. E. Woods, Digital Image Processing, 3rd ed., Pearson Education International, Pearson International Edition. (2008). [Online]. Available: https://sde.uoc.ac.in/sites/default/files/sde_videos/Digital%20Image%20Processing%203rd%20ed.%20-%20R.%20Gonzalez,%20R.%20Woods-ilovepdf-compressed.pdf. (Accessed: Mar. 23, 2026)
  37. J.-P. Fortin et al., "Harmonization of multi-site diffusion tensor imaging data," NeuroImage, vol. 161, pp. 149–170, (2017). http://doi.org/10.1016/j.neuroimage.2017.08.047
  38. L. G. Nyúl, J. K. Udupa, and X. Zhang, “New variants of a method of MRI scale standardization,” IEEE Trans. Med. Imaging, vol. 19, no. 2, pp. 143–150, (2000). http://doi.org/10.1109/42.836373
  39. J.-Y. Zhu, T. Park, P. Isola, and A. A. Efros, "Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks," arXiv e-prints, (2017). http://doi.org/10.48550/arXiv.1703.10593
  40. M. Sokolova and G. Lapalme, "A systematic analysis of performance measures for classification tasks," Information Processing & Management, vol. 45, no. 4, pp. 427–437, (2009). http://doi.org/10.1016/j.ipm.2009.03.002
  41. Z. Wang and A. C. Bovik, "Mean squared error: Love it or leave it? A new look at Signal Fidelity Measures," IEEE Signal Processing Magazine, vol. 26, no. 1, pp. 98–117, (2009). http://doi.org/10.1109/MSP.2008.930649
  42. L. R. Dice, "Measures of the Amount of Ecologic Association Between Species," Ecology, vol. 26, no. 3, pp. 297–302, (1945). http://doi.org/10.2307/1932409
  43. K. J. Gorgolewski et al., "The brain imaging data structure, a format for organizing and describing outputs of neuroimaging experiments," Sci. Data, vol. 3, p. 160044, Jun. (2016). http://doi.org/10.1038/sdata.2016.44