A novel quantitative method to evaluate lumbar disc degeneration: MRI histogram analysis.

Area of Science:

• Radiology
• Medical Imaging Analysis
• Biomedical Engineering

Background:

• Lumbar disc degeneration (LDD) is a common condition affecting spinal health.
• Traditional assessment of LDD relies on subjective classifications like the Pfirrmann method.
• There is a need for objective and quantitative tools to assess LDD.

Purpose of the Study:

• To evaluate lumbar disc degeneration (LDD) using MRI histogram analysis on routine MRI sequences.
• To determine the correlation between MRI histogram analysis and the established Pfirrmann classification.
• To assess the objectivity and stability of MRI histogram analysis through comprehensive agreement studies.

Main Methods:

• Inclusion of lumbar MRI images from 133 subjects.
• Classification of LDD using Pfirrmann grades and measurement of peak separation values via MRI histogram analysis.
• Correlation analysis between the two methods, determination of cutoff values, and intraclass correlation coefficient (ICC) analysis for inter-resolution, inter-observer, inter-ROI, and inter-slice agreement.

Main Results:

• Strong correlation found between peak separation values and Pfirrmann grades (r = -0.847).
• Good to excellent inter-resolution agreement (ICCs 0.822-0.916) for compressed resolutions, except for the lowest resolution (ICC = 0.533).
• Excellent inter-observer (ICC = 0.982) and inter-ROI (ICC = 0.915) agreements, with good inter-slice agreements for adjacent slices.

Conclusions:

• MRI histogram analysis on routine sequences is strongly correlated with the Pfirrmann classification for LDD assessment.
• The method demonstrates good objectivity and stability across various agreement scenarios (inter-resolution, inter-observer, inter-ROI, inter-slice).
• MRI histogram analysis provides an effective, objective, and quantitative tool for evaluating LDD, expanding its clinical application.