Abstract
PURPOSE: This study aimed to evaluate the influence of different cone-beam computed tomography (CBCT) image reconstruction parameters (slice thickness, noise filter application and orthogonal plane) on the calculation of bone fractal dimension and, based on those findings, to determine the optimal protocol for this type of assessment. MATERIALS AND METHODS: The sample consisted of 18 patients who underwent CBCT scans of the mandible and bone densitometry examinations. Four mandibular regions of interest were selected from the scans, with various image reconstruction parameters applied. Fractal dimension was calculated using the box-counting method. Two independent observers performed the evaluations, and all analyses were conducted with a significance level of 5%. RESULTS: The retromolar triangle and mandibular body regions did not demonstrate statistically significant differences when different tomographic reconstruction parameters were applied (P>0.05). The mandibular base did not display a consistent pattern that could define the influence of these parameters on its evaluation. The symphysis region showed improved performance in fractal analysis when using sagittal plane images with a 1 mm slice thickness. CONCLUSION: Operator-dependent parameters inherent to navigation software can influence fractal dimension analysis, with variations depending on the region of interest. The most appropriate parameters for this evaluation were identified as the sagittal plane with a 1 mm slice thickness. Among the regions assessed, the mandibular body was found to be the most suitable for fractal dimension analysis in CBCT.