Page 124 - Craniomaxillofacial Implant Surgery - Jeroen P.J. Dings
P. 124

                                122
Chapter 7
The high inter- and intra-observer reliability, ranging between 0.98 and 0.99 for both CBCT and MDCT measurements, was comparable with the available literature3. However, our study showed less inter-observer variation in linear measurements on cross-sectional MDCT images, as compared to CBCT images.
The accuracy of linear measurements in this study may be influenced by several factors. A possible drawback of this study is the use of dry skull models without providing a soft tissue equivalent attenuation. However, the majority of the ‘in vitro’ studies assessing the accuracy of potential sites of implant placement use dry skulls, mandibles or maxillae3. Three-dimensional volumetric depictions depend upon appropriate segmentation by means of thresholding. Voxels residing on tissue boundaries may contain different tissue types. Erroneous allocation of voxels to ‘soft tissue’ instead of ‘bone’ may occur and is known as the partial volume effect leading to subsequent measurement error8. This process is dependent on the software algorithm, the spatial- and contrast resolution of the scan, the thickness and degree of calcification, or cortication, of the bony structure. Although literature shows that accuracy outcomes are similar with- and without soft tissues, clinical extrapolation of the findings from our study is suboptimal, as experimental conditions differ from clinical9,10.
Furthermore, head orientation and position during image acquisition may influence measurement accuracy. Although cadaver heads used in this study were positioned and stabilized, as in a real clinical situation the eccentric anatomical locations may affect linear measurements. The role of the position of the head on linear measurement accuracy is still controversial. Several studies found no significant difference in measurement value with regard to different head positions or inclinations11,12. In contrast to these findings, Sabban et al. (2015) described a significant effect of the head position on measurement reliability in CBCT scans on intra-oral locations13. A systematic literature search by Wismeijer et al. (2018) revealed no adverse effect of the size of the field of view and partial rotations (180° vs. 360°) on linear measurements2.
Another potential drawback in our study is the use of standard settings of image acquisition parameters for both MDCT as CBCT scanners. The voxel size on MDCT and CBCT images were 0.3 mm and 0.5 mm respectively. As there are multiple image acquisition protocols available for each MDCT and CBCT scanner, different procedures could have been considered. However, a systematic review in this subject area by Fokas et al. denied a relation between different voxel sizes and measurement accuracy3.





























































































   122   123   124   125   126