BACKGROUND: Disc degeneration quantification is important for monitoring the effects of new therapeutic methods, such as cell and growth factor therapy. Magnetic resonance (MR) image texture reflects biochemical and structural tissue properties and has been used for differentiating between normal and pathological status in a variety of medical applications. PURPOSE: To investigate the suitability of textural descriptors for the quantification of intervertebral disc degeneration using conventional T2-weighted magnetic resonance images of the lumbar spine. MATERIAL AND METHODS: A 3 Tesla scanner was used, and conventional T2- weighted MR images were obtained, and a total of 255 lumbar discs were analyzed. An atlas-based method was used for segmenting the disc regions from the images. A set of first and second order statistics describing texture of each region were calculated. The validity and reliability of these descriptors for disc degeneration severity quantification was tested through their correlation with patient age and qualitative clinical grading of degeneration severity. Texture quantification results were compared to a widely accepted method for disc degeneration quantification based on the measurement of disc's mean signal intensity. RESULTS: Out of the set of texture descriptors tested, two descriptors quantifying image intensity inhomogeneity, i.e. the grey level standard deviation and co-occurrence derived sum of squares displayed the strongest association to patient age and clinical grading of disc degeneration severity (P<0.001). This is attributed to these inhomogeneity descriptors' capability to capture the progressive loss of nucleus-annulus distinction in the degenerative progress. Statistical analysis indicates that these descriptors can effectively separate between early stages of degeneration. Quantitative measurements are highly repeatable (intraclass correlation >0.98). CONCLUSION: Inhomogeneity descriptors could be a valuable tool for tracking the evolution of disc degeneration and monitoring the response to treatment in a simple, precise and repeatable manner.
BACKGROUND:Disc degeneration quantification is important for monitoring the effects of new therapeutic methods, such as cell and growth factor therapy. Magnetic resonance (MR) image texture reflects biochemical and structural tissue properties and has been used for differentiating between normal and pathological status in a variety of medical applications. PURPOSE: To investigate the suitability of textural descriptors for the quantification of intervertebral disc degeneration using conventional T2-weighted magnetic resonance images of the lumbar spine. MATERIAL AND METHODS: A 3 Tesla scanner was used, and conventional T2- weighted MR images were obtained, and a total of 255 lumbar discs were analyzed. An atlas-based method was used for segmenting the disc regions from the images. A set of first and second order statistics describing texture of each region were calculated. The validity and reliability of these descriptors for disc degeneration severity quantification was tested through their correlation with patient age and qualitative clinical grading of degeneration severity. Texture quantification results were compared to a widely accepted method for disc degeneration quantification based on the measurement of disc's mean signal intensity. RESULTS: Out of the set of texture descriptors tested, two descriptors quantifying image intensity inhomogeneity, i.e. the grey level standard deviation and co-occurrence derived sum of squares displayed the strongest association to patient age and clinical grading of disc degeneration severity (P<0.001). This is attributed to these inhomogeneity descriptors' capability to capture the progressive loss of nucleus-annulus distinction in the degenerative progress. Statistical analysis indicates that these descriptors can effectively separate between early stages of degeneration. Quantitative measurements are highly repeatable (intraclass correlation >0.98). CONCLUSION: Inhomogeneity descriptors could be a valuable tool for tracking the evolution of disc degeneration and monitoring the response to treatment in a simple, precise and repeatable manner.
Authors: A Neubert; J Fripp; C Engstrom; D Walker; M-A Weber; R Schwarz; S Crozier Journal: J Am Med Inform Assoc Date: 2013-06-27 Impact factor: 4.497
Authors: R Menezes-Reis; C E G Salmon; C S Carvalho; G P Bonugli; C B Chung; M H Nogueira-Barbosa Journal: AJNR Am J Neuroradiol Date: 2014-10-16 Impact factor: 3.825