Page 41 - Craniomaxillofacial Implant Surgery - Jeroen P.J. Dings

Reliability and accuracy of image-guided craniofacial implant planning
of available spatial resolution negatively influenced by motion blur and scatter of tissue.11,39
In this study, no relationship was found between accuracy of linear measurements and 2 different anatomical locations with varying bone widths. Representation of thin bone
layers on digital images may be reduced if bony densities are lowered with surrounding
air segmentation.3,20,21 Appropriate segmentation is essential in determining the
thresholding of bone pixel values and suppression of surrounding tissue values to enhance the structure of interest.28,44 Segmentation accuracy is influenced by the disadvantages of CBCT as scattered radiation, truncated view artifact and artifacts caused by beam hardening.10 ‘Partial averaging effect’ described by Gerlach et al3 may also hamper correct volumetric depiction of bony contours in averaging different densities within a voxel.
Diagnostic accuracy of CBCT is found to exhibit differences in relation to the anatomical location.12,28 Lascala et al47 found statistically significant underestimation of real measurements recorded at the skull base but not in dento-maxillofacial structures. Halperin48 et al indicated less diagnostic accuracy for the anterior areas compared to the posterior arches. No clear relationship between accuracy of measurements and anterior or posterior anatomical regions could be established in our study.
Lund et al22 showed that small distortions can result from the different reformation processes. In our study one reconstruction protocol was used enabling appropriate comparison of the different imaging modalities. The orientation of a radiographic plane parallel to the cut surface ensuring reliable measurements may have influenced measurements due to the slice-thickness which was set on a reconstructed slice increment of 0.3 mm for respectively CBCT and MDCT imaging. If this plane was not optimally positioned, an error in measurement is introduced due to possible variation in bony dimensions and mathematical reconstruction of images. The influence of varying slice thicknesses was not examined separately in this study. Literature shows an overestimation of distance measurements on CBCT and MDCT images resulting from increased slice thickness with margins of cortical bone appearing thicker.2,9
All measurements were executed on the same LCD monitors with identical mouse sensitivity. Al-Ekrish et al49 mentioned no differences in the reliability of linear measurements using different LCD monitors and different contrast resolution
39