Automatic Paraspinal Muscle Segmentation in Patients with Lumbar Pathology Using Deep Convolutional Neural Network

  • Wenyao XiaEmail author
  • Maryse Fortin
  • Joshua Ahn
  • Hassan Rivaz
  • Michele C. Battié
  • Terry M. Peters
  • Yiming Xiao
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11765)


Recent evidence suggests an association between low back pain (LBP) and changes in lumbar paraspinal muscle morphology and composition (i.e., fatty infiltration). Quantitative measurements of muscle cross-sectional areas (CSAs) from MRI scans are commonly used to examine the relationship between paraspinal muscle characters and different lumbar conditions. The current investigation primarily uses manual segmentation that is time-consuming, laborious, and can be inconsistent. However, no automatic MRI segmentation algorithms exist for pathological data, likely due to the complex paraspinal muscle anatomy and high variability in muscle composition among patients. We employed deep convolutional neural networks using U-Net+CRF-RNN with multi-data training to automatically segment paraspinal muscles from T2-weighted MRI axial slices at the L4-L5 and L5-S1 spinal levels and achieved averaged Dice score of 93.9\(\%\) and mean boundary distance of 1 mm. We also demonstrate the application using the segmentation results to reveal tissue characteristics of the muscles in relation to age and sex.


Segmentation Deep learning Lumbar pathologies MRI 



This work was supported by CIHR, CFI, NSERC and BrainsCAN, as well as the Seventh Framework Programme (Health-2007-2013, grant agreement NO: 201626: GENODISC) and Canada Reearch Chairs program. We acknowledge the support of NVIDIA Corporation and thank Dr. Yingli Lu for his help.


  1. 1.
    Balagué, F., Mannion, A.F., Pellisé, F., Cedraschi, C.: Non-specific low back pain. Lancet 379(9814), 482–491 (2012)CrossRefGoogle Scholar
  2. 2.
    Kamiya, N., Li, J., Kume, M., et al.: Fully automatic segmentation of paraspinal muscles from 3D torso CT images via multi-scale iterative random forest classifications. Int. J. Comput. Assist. Radiol. Surg. 13(11), 1697–1706 (2018)CrossRefGoogle Scholar
  3. 3.
    Engstrom, C.M., Fripp, J., Jurcak, V., et al.: Segmentation of the quadratus lumborum muscle using statistical shape modeling. J. Magn. Reson. Imaging 33(6), 1422–1429 (2011)CrossRefGoogle Scholar
  4. 4.
    Wei, Y., Xu, B., Tao, X., Qu, J.: Paraspinal muscle segmentation in CT images using a single atlas. In: Proceedings of the PIC, pp. 211–215. IEEE (2015)Google Scholar
  5. 5.
    Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. In: Navab, N., Hornegger, J., Wells, W.M., Frangi, A.F. (eds.) MICCAI 2015. LNCS, vol. 9351, pp. 234–241. Springer, Cham (2015). Scholar
  6. 6.
    Zheng, S., Jayasumana, S., Romera-Paredes, B., et al.: Conditional random fields as recurrent neural networks. In: ICCV, pp. 1529–1537 (2015)Google Scholar
  7. 7.
    Tustison, N.J., Avants, B.B., Cook, P.A., et al.: N4ITK: improved N3 bias correction. IEEE Trans. Med. Imaging 29(6), 1310 (2010)CrossRefGoogle Scholar
  8. 8.
    Xiao, Y., Fortin, M., Battié, M.C., Rivaz, H.: Population-averaged MRI atlases for automated image processing and assessments of lumbar paraspinal muscles. Eur. Spine J. 27(10), 2442–2448 (2018)CrossRefGoogle Scholar
  9. 9.
    Isensee, F., Petersen, J., Klein, A., et al.: nnU-Net: self-adapting framework for u-Net-based medical image segmentation. arXiv preprint arXiv:1809.10486 (2018)
  10. 10.
    Gibson, E., Li, W., Sudre, C., et al.: NiftyNet: a deep-learning platform for medical imaging. Comput. Methods Programs Biomed. 158, 113–122 (2018)CrossRefGoogle Scholar
  11. 11.
    Urrutia, J., Besa, P., Lobos, D., et al.: Lumbar paraspinal muscle fat infiltration is independently associated with sex, age, and inter-vertebral disc degeneration in symptomatic patients. Skeletal Radiol. 47(7), 955–961 (2018)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2019

Authors and Affiliations

  • Wenyao Xia
    • 1
    Email author
  • Maryse Fortin
    • 2
    • 3
  • Joshua Ahn
    • 4
  • Hassan Rivaz
    • 3
    • 5
  • Michele C. Battié
    • 6
  • Terry M. Peters
    • 1
  • Yiming Xiao
    • 1
  1. 1.Robarts Research InstituteWestern UniversityLondonCanada
  2. 2.Health, Kinesiology and Applied PhysiologyConcordia UniversityMontrealCanada
  3. 3.PERFORM CentreConcordia UniversityMontrealCanada
  4. 4.Department of KinesiologyWestern UniversityLondonCanada
  5. 5.Electrical and Computer EngineeringConcordia UniversityMontrealCanada
  6. 6.School of Physical Therapy and Western’s Bone and Joint InstituteWestern UniversityLondonCanada

Personalised recommendations