Methods: Sections of human mandible obtained from 7 individual cadavers (5 males and 2 females, avg=81.2 yrs) were scanned using cone beam computed tomography (CBCT) at the 200, 300 and 400 micron resolutions that are the same scanning range in clinic. Enamel and bone voxels in CBCT images were segmented from non-bone voxels. The alveolar bone (AB) region was digitally isolated within 1 mm outside root surface and the basal control bone (CB) was determined at 0.6 mm inside mandibular bone borders. The DBM histogram was obtained by calibrating a CT attenuation value of each bone voxel to a mineral density using known phantoms. Mean, standard deviation (SD), and coefficient of variation (COV=SD/mean) of DBM in each region were obtained. F-test for inter-specimen variances and paired t-tests for regional variations of the DBM parameters were performed.
Results: The means of DBM in CB had the least inter-CBCT variance resulting in significant differences from those in enamel at 300 and 400 micron scans (p<0.02). The SD and COV in AB were significant higher than those in CB independent of scanning resolutions (p<0.002).
Conclusions: The DBM in CB can likely be used as an internal reference for comparisons of DBM. The higher variability (SD and COV) of DBM in AB than that in CB would result from active bone remodeling due to masticatory functional demand. The regional variations of DBM were consistently assessed at all scanning resolutions.
Keywords: Bone, Digital image analysis, Enamel, Micro-CT and Remodeling