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

Vahid Abdollah; Eric C Parent; Samin Dolatabadi; Erica Marr; Keith Wachowicz; Michele Battié

doi:10.1007/s00586-021-07036-3

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. 2022 Aug;31(8):1979-1991. doi: 10.1007/s00586-021-07036-3. Epub 2021 Oct 30.

Authors

Vahid Abdollah^{1

2}, Eric C Parent³, Samin Dolatabadi⁴, Erica Marr⁴, Keith Wachowicz^{5

6}, Michele Battié^{7

8}

Affiliations

¹ Department of Physical Therapy, Faculty of Rehabilitation Medicine, University of Alberta, 2-50 Corbett Hall, Edmonton, AB, T6G2G4, Canada.
² 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.
³ Department of Physical Therapy, Faculty of Rehabilitation Medicine, University of Alberta, 2-50 Corbett Hall, Edmonton, AB, T6G2G4, Canada. eparent@ualberta.ca.
⁴ Faculty of Science, University of Alberta, 32 Athabasca Hall, Edmonton, AB, T6G 2E8, Canada.
⁵ Department of Oncology, Medical Physics Division, University of Alberta, 11560, Edmonton, AB, T6G 1Z2, Canada.
⁶ Department of Medical Physics, Cross Cancer Institute, 11560 University Avenue, Edmonton, AB, T6G 1Z2, Canada.
⁷ 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.
⁸ Faculty of Rehabilitation Medicine, University of Alberta, 2-50 Corbett Hall, Edmonton, AB, T6G 2G4, Canada.

PMID: 34718864
DOI: 10.1007/s00586-021-07036-3

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 (P_2-Offset, P_4-Offset) and information measure of correlation 1 (P_3-Offset, P_4-Offset) detected significant changes in the nucleus. Statistically significant changes were also observed for homogeneity (P_2-Offset, P_3-Offset), contrast (P_2-Offset), and difference variance (P_4-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.

Keywords: Compression; Intervertebral disc; Machine learning; Texture analysis; Traction.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Cross-Sectional Studies
Humans
Intervertebral Disc* / pathology
Lumbar Vertebrae* / pathology
Machine Learning
Magnetic Resonance Imaging / methods
Weight-Bearing / physiology