how accurate are the fusion of cone-beam ct and 3-d stereophotographic images如何准确cone-beam ct和三维stereophotographic图像的融合.pdfVIP

How Accurate Are the Fusion of Cone-Beam CT and 3-D Stereophotographic Images? 1 2 1 Yasas S. N. Jayaratne , Colman P. J. McGrath , Roger A. Zwahlen * 1 Discipline of Oral Maxillofacial Surgery, Faculty of Dentistry, The University of Hong Kong, Hong Kong, Hong Kong SAR, 2 Discipline of Periodontology and Public Health, Faculty of Dentistry, The University of Hong Kong, Hong Kong, Hong Kong SAR Abstract Background: Cone-beam Computed Tomography (CBCT) and stereophotography are two of the latest imaging modalities available for three-dimensional (3-D) visualization of craniofacial structures. However, CBCT provides only limited information on surface texture. This can be overcome by combining the bone images derived from CBCT with 3-D photographs. The objectives of this study were 1) to evaluate the feasibility of integrating 3-D Photos and CBCT images 2) to assess degree of error that may occur during the above processes and 3) to identify facial regions that would be most appropriate for 3-D image registration. Methodology: CBCT scans and stereophotographic images from 29 patients were used for this study. Two 3-D images corresponding to the skin and bone were extracted from the CBCT data. The 3-D photo was superimposed on the CBCT skin image using relatively immobile areas of the face as a reference. 3-D colour maps were used to assess the accuracy of superimposition were distance differences between the CBCT and 3-D photo were recorded as the signed average and the Root Mean Square (RMS) error. Principal Findings:: The signed average and RMS of the distance differences between the registered surfaces were 20.018 (60.129) mm and 0.739 ( 60.239) mm respectively. The m