Methods: Periodontal breakdown was induced by submerging a silk ligature on the maxillary second molars of Sprague-Dawley rats, and the harvested specimens was assessed by micro-CT and histology. A 0.4x0.8 mm box-shaped area adjacent to tooth surface was selected as the region of interest (ROI) from the equivalent slice of both images, and micro-CT threshold was generated from global iterative method (GI), global non-iterative method (GN), local edge specification (LE), and radiographic phantoms (RP), respectively for image segmentation. Bone ratio within ROI was compared between micro-CT and histology. Intra-examiner and intra-assessment reliability were compared for methodology validation.
Results: The results demonstrated high intra-examiner reliability in both micro-CT imaging and histomorphometry with high intraclass correlation coefficient (ICC>0.97) and low coefficient of variation (CV<5%). Images segmented by GI, GN, and LE, demonstrated a close correlation with histomorphometry (ICC>0.98), and threshold generated from local edge specification showed the highest ICC (0.994) and lowest CV (4.06%) than others.
Conclusions: The micro-CT imaging offers reliable mineral tissue measurements during periodontal breakdown, and local edge specification method is recommended for thresholding the image. The study is supported by NUS-FRC and NMRC-NIG research grants in Singapore.