Automated threshold-independent cortex segmentation by 3D-texture analysis of HR-pQCT scans

• Published in: Bone (New York, N.Y.) Vol. 51; no. 3; pp. 480 - 487
• Main Authors: Valentinitsch, Alexander, Patsch, Janina M., Deutschmann, Julia, Schueller-Weidekamm, Claudia, Resch, Heinrich, Kainberger, Franz, Langs, Georg
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier Inc 01.09.2012; Elsevier
• Subjects: Aged; Aged, 80 and over; Algorithms; Automation; Biological and medical sciences; Cortical bone metrics; Female; Fully automated segmentation; Fundamental and applied biological sciences. Psychology; HR-pQCT; Humans; Imaging, Three-Dimensional - methods; Male; Middle Aged; Organ Size; Orthopedics; Radius - anatomy & histology; Radius - diagnostic imaging; Regression Analysis; Texture analysis; Threshold-independent; Tissue Donors; Tomography, X-Ray Computed - methods; Vertebrates: anatomy and physiology, studies on body, several organs or systems; Texture analysis; HR-pQCT; CRTX; GT; SA; GLCM; TIST; Threshold-independent; PMMA; MLSE; Fully automated segmentation; Cortical bone metrics; DC; Manufacturer's standard algorithm; ground truth; Threshold-independent segmentation tool; Dice coefficient; Mean local segmentation error; High-resolution peripheral quantitative computed tomography; Segmentation algorithm by Burghardt et al.; Gray level co-occurrence matrix; Polymethylmethacrylate; Segmentation; Image processing; Morphology; Texture; Cortical bone
• ISSN: 8756-3282; 1873-2763; 1873-2763
• DOI: 10.1016/j.bone.2012.06.005

The quantitative assessment of metabolic bone diseases relies on tissue properties such as bone mineral density (BMD) and bone microarchitecture. In spite of an increasing number of publications using high-resolution peripheral quantitative computed-tomography (HR-pQCT), the accurate and reproducible separation of cortical and trabecular bone remains challenging. In this paper, we present a novel, fully automated, threshold-independent technique for the segmentation of cortical and trabecular bone in HR-pQCT scans. This novel post-processing method is based on modeling appearance characteristics from manually annotated cases. In our experiments the algorithm automatically selected texture features with high differentiating power and trained a classifier to separate cortical and trabecular bone. From this mask, cortical thickness and tissue volume could be calculated with high accuracy. The overlap between the proposed threshold-independent segmentation tool (TIST) and manual contouring was 0.904±0.045 (Dice coefficient). In our experiments, TIST obtained higher overall accuracy in our measurements than other techniques. ► Fully-automated segmentation of cortical and trabecular bone in HR-pQCT scans. ► Threshold-independent segmentation tool (TIST) based on 3D texture analysis. ► We provide full transparency of algorithm development.