A comparison of the thresholding strategies of micro-CT for periodontal bone loss: a pilot study.

OBJECTIVES: Micro-CT provides three-dimensional details and has been widely used for biomedical assessments. This study aimed to determine the most appropriate threshold method for quantitatively assessing the dynamics of periodontal destruction.

METHODS: Inflammation was induced by submerging a silk ligature in the sulcus of the maxillary second molars of rats, and the animals were killed prior to ligature placement and after 7 and 21 days. The maxillae were examined for the bone resorptive activities by micro-CT, histology and tartrate-resistant acid phosphatase staining. The imaging threshold was determined by CT phantom, global and local algorithms. A bone fraction measurement from each threshold-determining technique was compared with histomorphometry. The reliability and reproducibility were examined by the intraclass correlation coefficient (ICC) and the coefficient of variation.

RESULTS: Significant reduction of inflammatory infiltration (p < 0.01) and active osteoclastic resorption (p < 0.05) from Day 7 to Day 21 were noted. High inter- and intraexaminer agreement were demonstrated in both histomorphometric and micro-CT assessments (ICC > 0.98). The algorithm-based technique demonstrated stronger correlation to histomorphometry than phantom-based thresholds, and the highest agreement was presented by the local algorithm (ICC > 0.96). This, however, was considerably computationally expensive.

CONCLUSIONS: The local threshold-determining algorithm is suggested for examining inflammation-induced bone loss. Further investigation will be aimed at enhancing computational efficiency.