PurposeSuperimposition of two dimensional preoperative and postoperative facial images, including radiographs and photographs, are used to evaluate the surgical changes after orthognathic surgery. Recently, three dimensional (3D) imaging has been introduced allowing more accurate analysis of surgical changes. Surface based registration and voxel based registration are commonly used methods for 3D superimposition. The aim of this study was to evaluate and compare the accuracy of the two methods.Materials and methodsPre-operative and 6 months post-operative cone beam CT scan (CBCT) images of 31 patients were randomly selected from the orthognathic patient database at the Dental Hospital and School, University of Glasgow, UK. Voxel based registration was performed on the DICOM images (Digital Imaging Communication in Medicine) using Maxilim software (Medicim-Medical Image Computing, Belgium). Surface based registration was performed on the soft and hard tissue 3D models using VRMesh (VirtualGrid, Bellevue City, WA). The accuracy of the superimposition was evaluated by measuring the mean value of the absolute distance between the two 3D image surfaces. The results were statistically analysed using a paired Student t-test, ANOVA with post-hoc Duncan test, a one sample t-test and Pearson correlation coefficient test.ResultsThe results showed no significant statistical difference between the two superimposition methods (p<0.05). However surface based registration showed a high variability in the mean distances between the corresponding surfaces compared to voxel based registration, especially for soft tissue. Within each method there was a significant difference between superimposition of the soft and hard tissue models.ConclusionsThere were no significant statistical differences between the two registration methods and it was unlikely to have any clinical significance. Voxel based registration was associated with less variability. Registering on the soft tissue in isolation from the hard tissue may not be a true reflection of the surgical change.

No abstract

Abstract:The accuracy of three dimensional predictions of soft tissue changes following surgical correction of facial asymmetry has been evaluated in this study. The preoperative (T1) and 6-12 months postoperative (T2) CBCT scans of 13 patients were studied. All patients had surgical correction for facial asymmetry as part of the multidisciplinary treatment protocol.The magnitude of surgical movement was measured; virtual surgery was carried out on the preoperative scans using Maxilim software. The predicted soft tissue changes were compared to the actual postoperative appearance (T2) at specific anatomic facial regions. Mean (signed) distances and the mean (absolute) distances between the predicted and actual 3D surface meshes for each region were calculated. One sample Student t-test was applied. Additionally, a novel directional soft tissue analysis was applied to analyse the accuracy of the prediction of soft tissue. The results showed that the distances between the predicted and actual postoperative soft tissue changes were within 2.0mm (p<0.05) at all regions. The predicted facial morphology was narrower than the actual surgical changes at the cheek regions. Three dimensional soft tissue prediction for the correction of asymmetric cases is clinically acceptable using Maxilim software. Introduction:

No abstract