Comparison of the accuracy of voxel based registration and surface based registration for 3D assessment of surgical change following orthognathic surgery
Autor: | Almukhtar, A, Ayoub, F, Zuo, X, McDonald, J, Ju, X, Khambay, BS |
---|---|
Jazyk: | angličtina |
Rok vydání: | 2014 |
Předmět: |
Cone beam computed tomography
Image Processing Radiography medicine.medical_treatment Orthognathic surgery lcsh:Medicine computer.software_genre Diagnostic Radiology Voxel Medicine and Health Sciences Postoperative Period lcsh:Science Tomography Orthodontics Multidisciplinary Radiology and Imaging Orthognathic Surgery Digital imaging Software Engineering Cone-Beam Computed Tomography Research Design Preoperative Period Physical Sciences symbols Engineering and Technology Plastic Surgery and Reconstructive Techniques Statistics (Mathematics) Research Article Computer and Information Sciences medicine.medical_specialty Clinical Research Design Oral Medicine Surgical and Invasive Medical Procedures Biostatistics Research and Analysis Methods symbols.namesake DICOM Imaging Three-Dimensional Diagnostic Medicine Computer Graphics medicine Humans Superimposition Medical physics Retrospective Studies Software Tools business.industry lcsh:R Computing Methods Pearson product-moment correlation coefficient Computed Axial Tomography Face Dentistry Signal Processing lcsh:Q business computer Mathematics |
Zdroj: | PLoS ONE PLoS ONE, Vol 9, Iss 4, p e93402 (2014) |
ISSN: | 1932-6203 |
Popis: | Purpose:\ud \ud Superimposition 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.\ud \ud Materials and methods:\ud \ud Pre-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.\ud \ud Results:\ud \ud The results showed no significant statistical difference between the two superimposition methods (p\ud \ud Conclusions:\ud \ud There 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. |
Databáze: | OpenAIRE |
Externí odkaz: |