BACKGROUND: Segmentation of the gray and white matter (GM, WM) of the human spinal cord in MRI images as well as the analysis of spinal cord diffusivity are challenging. When appropriately segmented, diffusion tensor imaging (DTI) of the spinal cord might be beneficial in the diagnosis and prognosis of several diseases. PURPOSE: To evaluate the applicability of a semiautomatic algorithm provided by ITK-SNAP in classification mode (CLASS) for segmenting cervical spinal cord GM, WM in MRI images and analyzing DTI parameters. STUDY TYPE: Prospective. SUBJECTS: Twenty healthy volunteers. SEQUENCES: 1.5T, turbo spin echo, fast field echo, single-shot echo planar imaging. ASSESSMENT: Three raters segmented the tissues by manual, CLASS, and atlas-based methods (Spinal Cord Toolbox, SCT) on T2 -weighted and DTI images. Masks were quantified by similarity and distance metrics, then analyzed for repeatability and mutual comparability. Masks created over T2 images were registered into diffusion space and fractional anisotropy (FA) values were statistically evaluated for dependency on method, rater, or tissue. STATISTICAL TESTS: t-test, analysis of variance (ANOVA), coefficient of variation, Dice coefficient, Hausdorff distance. RESULTS: CLASS segmentation reached better agreement with manual segmentation than did SCT (P < 0.001). Intra- and interobserver repeatability of SCT was better for GM and WM (both P < 0.001) but comparable with CLASS in entire spinal cord segmentation (P = 0.17 and P = 0.07, respectively). While FA values of whole spinal cord were not influenced by choice of segmentation method, both semiautomatic methods yielded lower FA values (P < 0.005) for GM than did the manual technique (mean differences 0.02 and 0.04 for SCT and CLASS, respectively). Repeatability of FA values for all methods was sufficient, with mostly less than 2% variance. DATA CONCLUSION: The presented semiautomatic method in combination with the proposed approach to data registration and analyses of spinal cord diffusivity can potentially be used as an alternative to atlas-based segmentation. LEVEL OF EVIDENCE: 1 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;47:1217-1227.

Department of Biophysics Faculty of Medicine Masaryk University Brno Czech Republic

Department of Biophysics Faculty of Medicine Masaryk University Brno Czech Republic Department of Radiology University Hospital Brno and Masaryk University Brno Czech Republic

Department of Radiology University Hospital Brno and Masaryk University Brno Czech Republic

Institute of Biostatistics and Analyses Faculty of Medicine Masaryk University Brno Czech Republic

References provided by Crossref.org