Skip to main content

Advertisement

SpringerLink
Log in

Use of machine learning to select texture features in investigating the effects of axial loading on T2-maps from magnetic resonance imaging of the lumbar discs

  • Original Article
  • Published:
European Spine Journal Aims and scope Submit manuscript

Abstract

Background

Recent advances in texture analysis and machine learning offer new opportunities to improve the application of imaging to intervertebral disc biomechanics. This study employed texture analysis and machine learning on MRIs to investigate the lumbar disc’s response to loading.

Methods

Thirty-five volunteers (30 (SD 11) yrs.) with and without chronic back pain spent 20 min lying in a relaxed unloaded supine position, followed by 20 min loaded in compression, and then 20 min with traction applied. T2-weighted MR images were acquired during the last 5 min of each loading condition. Custom image analysis software was used to segment discs from adjacent tissues semi-automatically and segment each disc into the nucleus, anterior and posterior annulus automatically. A grey-level, co-occurrence matrix with one to four pixels offset in four directions (0°, 45°, 90° and 135°) was then constructed (320 feature/tissue). The Random Forest Algorithm was used to select the most promising classifiers. Linear mixed-effect models and Cohen’s d compared loading conditions.

Findings

All statistically significant differences (p < 0.001) were observed in the nucleus and posterior annulus in the 135° offset direction at the L4-5 level between lumbar compression and traction. Correlation (P2-Offset, P4-Offset) and information measure of correlation 1 (P3-Offset, P4-Offset) detected significant changes in the nucleus. Statistically significant changes were also observed for homogeneity (P2-Offset, P3-Offset), contrast (P2-Offset), and difference variance (P4-Offset) of the posterior annulus.

Interpretation

MRI textural features may have the potential of identifying the disc's response to loading, particularly in the nucleus and posterior annulus, which appear most sensitive to loading.

Level of evidence

Diagnostic: individual cross-sectional studies with consistently applied reference standard and blinding

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2

Similar content being viewed by others

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Code availability

Image processing software is freely available at https://github.com/vabdollah/SpineGUI).

References

  1. Jinkins JR, Dworkin JS, Damadian RV (2005) Upright, weight-bearing, dynamic-kinetic MRI of the spine: initial results. Eur Radiol 15:1815–1825. https://doi.org/10.1007/s00330-005-2666-4

    Article  PubMed  Google Scholar 

  2. Hebelka H, Torén L, Lagerstrand K, Brisby H (2018) Axial loading during MRI reveals deviant characteristics within posterior IVD regions between low back pain patients and controls. Eur Spine J 27:2840–2846. https://doi.org/10.1007/s00586-018-5774-y

    Article  CAS  PubMed  Google Scholar 

  3. Wang Y-C, Jeng C-M, Wu C-Y et al (2008) Dynamic effects of axial loading on the lumbar spine during magnetic resonance imaging in patients with suspected spinal stenosis. J Formosan Med Associ = Taiwan yi zhi. https://doi.org/10.1016/S0929-6646(08)60095-9

    Article  Google Scholar 

  4. Nilsson M, Lagerstrand K, Kasperska I et al (2016) Axial loading during MRI influences T2-mapping values of lumbar discs: a feasibility study on patients with low back pain. Eur Spine J 25:2856–2863. https://doi.org/10.1007/s00586-016-4670-6

    Article  PubMed  Google Scholar 

  5. Abdollah V, Parent EC, Battié MC (2018) MRI evaluation of the effects of extension exercises on the disc fluid content and location of the centroid of the fluid distribution. Musculoskel Sci Pract 33:67–70. https://doi.org/10.1016/j.msksp.2017.11.008

    Article  Google Scholar 

  6. Hirasawa Y, Bashir WA, Smith FW et al (2007) Postural changes of the dural sac in the lumbar spines of asymptomatic individuals using positional stand-up magnetic resonance imaging. Spine 32:E136–E140. https://doi.org/10.1097/01.brs.0000255202.94153.ca

    Article  PubMed  Google Scholar 

  7. Mauch F, Jung C, Huth J, Bauer G (2010) Changes in the lumbar spine of athletes from supine to the true-standing position in magnetic resonance imaging. Spine 35:1002–1007. https://doi.org/10.1097/BRS.0b013e3181bdb2d3

    Article  PubMed  Google Scholar 

  8. Kanno H, Ozawa H, Koizumi Y et al (2012) Dynamic change of dural sac cross-sectional area in axial loaded magnetic resonance imaging correlates with the severity of clinical symptoms in patients with lumbar spinal canal stenosis. Spine 37:207–213. https://doi.org/10.1097/BRS.0b013e3182134e73

    Article  PubMed  Google Scholar 

  9. Davies-Tuck ML, Wluka AE, Forbes A et al (2010) Development of bone marrow lesions is associated with adverse effects on knee cartilage while resolution is associated with improvement–a potential target for prevention of knee osteoarthritis: a longitudinal study. Arthritis Res Ther 12:R10. https://doi.org/10.1186/ar2911

    Article  PubMed  PubMed Central  Google Scholar 

  10. Nazari J, Pope MH, Graveling RA (2012) Reality about migration of the nucleus pulposus within the intervertebral disc with changing postures. Clin Biomech (Bristol, Avon) 27:213–217. https://doi.org/10.1016/j.clinbiomech.2011.09.011

    Article  Google Scholar 

  11. Karadimas EJ, Siddiqui M, Smith FW, Wardlaw D (2006) Positional MRI changes in supine versus sitting postures in patients with degenerative lumbar spine. J Spinal Disord Tech 19:495–500. https://doi.org/10.1097/01.bsd.0000211213.98070.c2

    Article  PubMed  Google Scholar 

  12. Lee S-U, Hargens AR, Fredericson M, Lang PK (2003) Lumbar spine disc heights and curvature: upright posture vs. supine compression harness. Aviat Space Environ Med 74:512–516

    PubMed  Google Scholar 

  13. Iwata T, Miyamoto K, Hioki A et al (2013) In vivo measurement of lumbar foramen during axial loading using a compression device and computed tomography. J Spinal Disord Tech 26:E177–E182. https://doi.org/10.1097/BSD.0b013e318286f635

    Article  PubMed  Google Scholar 

  14. Abdollah V, Parent EC, Su A et al (2021) The effects of axial loading on the morphometric and T 2 characteristics of lumbar discs in relation to disc degeneration. Clin Biomech 83:105291. https://doi.org/10.1016/j.clinbiomech.2021.105291

    Article  Google Scholar 

  15. Stelzeneder D, Kovács BK, Goed S et al (2012) Effect of short-term unloading on T2 relaxation time in the lumbar intervertebral disc–in vivo magnetic resonance imaging study at 3.0 tesla. The spine journal : official Journal of the North American Spine Society 12:257–264. https://doi.org/10.1016/j.spinee.2012.02.001

    Article  Google Scholar 

  16. Hebelka H, Miron A, Kasperska I et al (2018) Axial loading during MRI induces significant T2 value changes in vertebral endplates—a feasibility study on patients with low back pain. J Orthop Surg Res 13:18. https://doi.org/10.1186/s13018-018-0727-z

    Article  PubMed  PubMed Central  Google Scholar 

  17. Borenstein D, O’Mara JJ, Boden S et al (1990) Abnormal magnetic-resonance scans of the lumbar spine in asymptomatic subjects. a prospective investigation. J Bone Joint Surg Am 72:403–408

    Article  Google Scholar 

  18. Kalichman L, Kim DH, Li L et al (2010) Computed tomography-evaluated features of spinal degeneration: prevalence, intercorrelation, and association with self-reported low back pain. Spine J 10:200–208. https://doi.org/10.1016/j.spinee.2009.10.018

    Article  PubMed  Google Scholar 

  19. Wiesel SW, Tsourmas N, Feffer HL et al (1984) A study of computer-assisted tomography I. The incidence of positive CAT scans in an asymptomatic group of patients. Spine 9:549–551. https://doi.org/10.1097/00007632-198409000-00003

    Article  CAS  PubMed  Google Scholar 

  20. Savage RA, Whitehouse GH, Roberts N (1997) The relationship between the magnetic resonance imaging appearance of the lumbar spine and low back pain, age and occupation in males. Eur Spine J 6:106–114. https://doi.org/10.1007/BF01358742

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Brinjikji W, Luetmer PH, Comstock B et al (2015) Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. AJNR Am J Neuroradiol 36:811–816. https://doi.org/10.3174/ajnr.A4173

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  22. Roberts BC, Perilli E, Reynolds KJ (2014) Application of the digital volume correlation technique for the measurement of displacement and strain fields in bone: a literature review. J Biomech 47:923–934

    Article  Google Scholar 

  23. Abdollah V, Parent EC, Dolatabadi S et al (2020) Texture analysis in the classification of T 2 -weighted magnetic resonance images in persons with and without low back pain. J Orthop Res. 39(10):2187–2196. https://doi.org/10.1002/jor.24930

    Article  CAS  PubMed  Google Scholar 

  24. Haralick RM, Dinstein I, Shanmugam K, Dinstein I (1973) Textural features for image classification. IEEE Trans Sys Man Cybern. https://doi.org/10.1109/TSMC.1973.4309314

    Article  Google Scholar 

  25. Chevrefils C, Cheriet F, Aubin CÉ, Grimard G (2009) Texture analysis for automatic segmentation of intervertebral disks of scoliotic spines from MR images. IEEE Trans Inf Technol Biomed 13(4):608–620

    Article  Google Scholar 

  26. Apostol L, Boudousq V, Basset O et al (2006) Relevance of 2D radiographic texture analysis for the assessment of 3D bone micro-architecture. Med Phys 33:3546–3556. https://doi.org/10.1118/1.2211727

    Article  PubMed  Google Scholar 

  27. Rachidi M, Marchadier A, Gadois C et al (2008) Laws’ masks descriptors applied to bone texture analysis: An innovative and discriminant tool in osteoporosis. Skeletal Radiol 37:541–548. https://doi.org/10.1007/s00256-008-0463-2

    Article  CAS  PubMed  Google Scholar 

  28. Ranjanomennahary P, Ghalila SS, Malouche D et al (2011) Comparison of radiograph-based texture analysis and bone mineral density with three-dimensional microarchitecture of trabecular bone. Med Phys 38:420–428. https://doi.org/10.1118/1.3528125

    Article  CAS  PubMed  Google Scholar 

  29. Muehlematter UJ, Mannil M, Becker AS et al (2019) Vertebral body insufficiency fractures: detection of vertebrae at risk on standard CT images using texture analysis and machine learning. Eur Radiol 29:2207–2217. https://doi.org/10.1007/s00330-018-5846-8

    Article  PubMed  Google Scholar 

  30. Michopoulou S, Costaridou L, Vlychou M et al (2011) Texture-based quantification of lumbar intervertebral disc degeneration from conventional T2-weighted MRI. Acta Radiol 52:91–98. https://doi.org/10.1258/ar.2010.100166

    Article  PubMed  Google Scholar 

  31. Soh LK, Tsatsoulis C (1999) Texture analysis of sar sea ice imagery using gray level co-occurrence matrices. IEEE Trans Geosci Remote Sens 37:780–795. https://doi.org/10.1109/36.752194

    Article  Google Scholar 

  32. Clausi DA (2002) An analysis of co-occurrence texture statistics as a function of grey level quantization. Can J Remote Sens 28:45–62. https://doi.org/10.5589/m02-004

    Article  Google Scholar 

  33. Portney LG (2020) Foundations of clinical research: applications to practice. F.A. Davis Co., NJ

    Google Scholar 

  34. Marsland S (2014) Machine learning. CRC Press, FL

    Book  Google Scholar 

  35. Breiman L (2001) Random forests. Mach Learn 45:5–32. https://doi.org/10.1023/A:1010933404324

    Article  Google Scholar 

  36. O’Connor JPB, Rose CJ, Waterton JC et al (2015) Imaging intratumor heterogeneity: role in therapy response, resistance, and clinical outcome. Clin Cancer Res 21:249–257

    Article  Google Scholar 

  37. Brinjikji W, Diehn FE, Jarvik JG et al (2015) MRI findings of disc degeneration are more prevalent in adults with low back pain than in asymptomatic controls: a systematic review and meta-analysis. Am J Neuroradiol 36:2394–2399. https://doi.org/10.3174/ajnr.A4498

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  38. Deyo RA, Dworkin SF, Amtmann D et al (2014) Report of the NIH task force on research standards for chronic low back pain. Pain medicine (Malden, Mass) 15:1249–1267. https://doi.org/10.1111/pme.12538;10.1111/pme.12538

    Article  Google Scholar 

  39. Nowicki BH, Yu S, Reinartz J et al (1990) Effect of axial loading on neural foramina and nerve roots in the lumbar spine. Radiology 176:433–437. https://doi.org/10.1148/radiology.176.2.2367657

    Article  CAS  PubMed  Google Scholar 

  40. Apfel CC, Cakmakkaya OS, Martin W et al (2010) Restoration of disk height through non-surgical spinal decompression is associated with decreased discogenic low back pain: a retrospective cohort study. BMC Musculoskelet Disord 11:155. https://doi.org/10.1186/1471-2474-11-155;10.1186/1471-2474-11-155

    Article  PubMed  PubMed Central  Google Scholar 

  41. Goh S, Tan C, Price RI et al (2000) Influence of age and gender on thoracic vertebral body shape and disc degeneration: an MR investigation of 169 cases. J Anat 197(Pt 4):647–657

    Article  Google Scholar 

  42. Abdollah V, Parent EC, Su A et al (2020) Could compression and traction loading improve the ability of magnetic resonance imaging to identify findings related to low back pain? Musculosk Sci Pract. https://doi.org/10.1016/j.msksp.2020.102250

    Article  Google Scholar 

  43. Pfirrmann CW, Metzdorf A, Zanetti M et al (2001) Magnetic resonance classification of lumbar intervertebral disc degeneration. Spine 26:1873–1878

    Article  CAS  Google Scholar 

  44. Abdollah V, Parent EC, Battié MC (2018a) Is the location of the signal intensity weighted centroid a reliable measurement of fluid displacement within the disc? Biomed Tech (Berl) 63:453–460

    Article  Google Scholar 

  45. Abdollah V, Parent EC, Battié MC (2018b) Reliability and validity of lumbar disc height quantification methods using magnetic resonance images. Biomed Tech (Berl) 64:111–117

    Google Scholar 

  46. Lagarias JC, Reeds JA, Wright MH, Wright PE (1998) Convergence properties of the nelder-mead simplex method in low dimensions. SIAM J Optim 9:112–147. https://doi.org/10.1137/S1052623496303470

    Article  Google Scholar 

  47. Ramola A, Kumar A, Shakya D, Pham van, (2020) Study of statistical methods for texture analysis and their modern evolutions. Eng Rep. https://doi.org/10.1002/eng2.12149

    Article  Google Scholar 

  48. Kuznetsova A, Brockhoff PB, Christensen RHB (2017) lmer test package: tests in linear mixed effects models. J Stat Software 82:1–26

    Article  Google Scholar 

  49. Espinoza Orías AA, Mammoser NM, Triano JJ et al (2016) Effects of axial torsion on disc height distribution: an in vivo study. J Manipulative Physiol Ther 39:294–303. https://doi.org/10.1016/j.jmpt.2016.03.002

    Article  PubMed  PubMed Central  Google Scholar 

  50. Owens SC, Brismee JM, Pennell PN et al (2009) Changes in spinal height following sustained lumbar flexion and extension postures: a clinical measure of intervertebral disc hydration using stadiometry. J Manipulative Physiol Ther 32:358–363. https://doi.org/10.1016/j.jmpt.2009.04.006;10.1016/j.jmpt.2009.04.006

    Article  PubMed  Google Scholar 

  51. Kimura S, Steinbach GC, Watenpaugh DE, Hargens AR (2001) Lumbar spine disc height and curvature responses to an axial load generated by a compression device compatible with magnetic resonance imaging. Spine 26:2596–2600

    Article  CAS  Google Scholar 

  52. Stelzeneder D, Welsch GH, Kovacs BK et al (2012) Quantitative T2 evaluation at 3.0T compared to morphological grading of the lumbar intervertebral disc: a standardized evaluation approach in patients with low back pain. Eur J Radiol 81:324–330. https://doi.org/10.1016/j.ejrad.2010.12.093;10.1016/j.ejrad.2010.12.093

    Article  PubMed  Google Scholar 

  53. Materka A (2004) Texture analysis methodologies for magnetic resonance imaging. Dialogues Clin Neurosci 6:243–250

    Article  Google Scholar 

  54. Herlidou-Même S, Constans JM, Carsin B et al (2003) MRI texture analysis on texture test objects, normal brain and intracranial tumors. Magn Reson Imaging 21:989–993. https://doi.org/10.1016/S0730-725X(03)00212-1

    Article  PubMed  Google Scholar 

  55. Zhao Q, Shi CZ, Luo LP (2014) Role of the texture features of images in the diagnosis of solitary pulmonary nodules in different sizes. China J Cancer Res 26:451–458. https://doi.org/10.3978/j.issn.1000-9604.2014.08.07

    Article  Google Scholar 

  56. Ramos G, Martin W (1994) Effects of vertebral axial decompression on intradiscal pressure. J Neurosurg 81:350–353. https://doi.org/10.3171/jns.1994.81.3.0350

    Article  CAS  PubMed  Google Scholar 

  57. Magnusson ML, Aleksiev AR, Spratt KF et al (1996) Hyperextension and spine height changes. Spine 21:2670–2675

    Article  CAS  Google Scholar 

  58. Gerke DA, Brismee JM, Sizer PS et al (2011) Change in spine height measurements following sustained mid-range and end-range flexion of the lumbar spine. Appl Ergon 42:331–336. https://doi.org/10.1016/j.apergo.2010.08.003;10.1016/j.apergo.2010.08.003

    Article  PubMed  Google Scholar 

  59. Madsen R, Jensen TS, Pope M et al (2008) The effect of body position and axial load on spinal canal morphology: an MRI study of central spinal stenosis. Spine 33:61–67. https://doi.org/10.1097/BRS.0b013e31815e395f

    Article  PubMed  Google Scholar 

  60. Kinder A, Filho FP, Ribeiro E et al (2012) Magnetic resonance imaging of the lumbar spine with axial loading: a review of 120 cases. Eur J Radiol 81:e561–e564. https://doi.org/10.1016/j.ejrad.2011.06.027

    Article  PubMed  Google Scholar 

  61. Wilke HJ, Neef P, Caimi M et al (1999) New in vivo measurements of pressures in the intervertebral disc in daily life. Spine 24:755–762

    Article  CAS  Google Scholar 

  62. Nachemson A (1960) Lumbar intradiscal pressure. Experimental studies on post-mortem material. Acta orthopaedica ScandinavicaSupplementum 43:1–104

    CAS  Google Scholar 

Download references

Funding

This study was funded by grants from Canadian Physiotherapy Association, Alberta Spine Foundation, and University of Alberta’s Faculty of Rehabilitation Medicine.

Author information

Author notes

  1. Vahid Abdollah

    Present address: Neuromuscular Control and Biomechanics Laboratory, Department of Mechanical Engineering, Faculty of Engineering, Nanotechnology Research Centre, University of Alberta, 5-046, Edmonton, AB, T6G 2M9, Canada

Authors and Affiliations

  1. Department of Physical Therapy, Faculty of Rehabilitation Medicine, University of Alberta, 2-50 Corbett Hall, Edmonton, AB, T6G2G4, Canada

    Vahid Abdollah & Eric C. Parent

  2. Faculty of Science, University of Alberta, 32 Athabasca Hall, Edmonton, AB, T6G 2E8, Canada

    Samin Dolatabadi & Erica Marr

  3. Department of Oncology, Medical Physics Division, University of Alberta, 11560, Edmonton, AB, T6G 1Z2, Canada

    Keith Wachowicz

  4. Department of Medical Physics, Cross Cancer Institute, 11560 University Avenue, Edmonton, AB, T6G 1Z2, Canada

    Keith Wachowicz

  5. Western Research Chair in Musculoskeletal Exercise, Mobility and Health, School of Physical Therapy and Western’s Bone & Joint Institute, Western University, London, ON, N6A 3K7, Canada

    Michele Battié

  6. Faculty of Rehabilitation Medicine, University of Alberta, 2-50 Corbett Hall, Edmonton, AB, T6G 2G4, Canada

    Michele Battié

Authors
  1. Vahid Abdollah
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Eric C. Parent
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Samin Dolatabadi
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Erica Marr
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Keith Wachowicz
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Michele Battié
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Eric C. Parent.

Ethics declarations

Conflicts of interest

The authors declare that they have no conflict of interest.

Ethics approval

The Health Research Ethics Board of the University of Alberta has approved the current study (Pro00052494).

Consent to participate

Informed consent was obtained from all individual participants included in the study.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Abdollah, V., Parent, E.C., Dolatabadi, S. et al. Use of machine learning to select texture features in investigating the effects of axial loading on T2-maps from magnetic resonance imaging of the lumbar discs. Eur Spine J 31, 1979–1991 (2022). https://doi.org/10.1007/s00586-021-07036-3

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00586-021-07036-3

Keywords

Search

Navigation