3-Dimensional Morphologic Analysis of the Craniofacial Skeletal Complex Using Geometry-Based Superimposition and Normalization
- Author(s): Sung, Jay Hyuck
- Advisor(s): Tetradis, Sotirios
- Moon, Won
- et al.
Introduction : Cephalometrics is a radiographic technique used to analyze craniofacial structures for diagnostic or analytic purposes. However, conventional cephalometrics has many limitations, including the fact that it is entirely landmark-dependent and based on a 2-dimensional image. In previous studies, Fourier descriptors have been used to describe part of the craniofacial structure. However, these studies are insufficient providing a complete set of information to fully analyze craniofacial morphology. The aim of this study is to develop a true 3-dimensional description of the craniofacial structure that will provide the basis of morphologic analysis in a more comprehensive and effective way.
Materials and Methods : CBCT images taken at UCLA School of Dentistry by the Newtom 3G CBCT scanner (Image Works) were collected. Samples with significant craniofacial defects were excluded. Using 10 samples without morphologic abnormality, the curved outline of the craniofacial structure was defined by collecting the coordinates of the points along the border of the shape. The superimposition and averaging was done using procrustes analysis. Geometric algebraic methods were used to combine all the curved outlines to construct an average shape of the whole craniofacial structure allowing for superimposition and comparison.
Results : The outline of the skull was successfully aligned in 3-dimensional space to represent the craniofacial structure. The normalized form provided a basis for the comparison of an individual sample to the group average. The new method showed advantages over the conventional cephalometrics by eliminating the constraint factors of 2-dimensional images. This allows us to analyze difference and irregularities of craniofacial morphology in a more accurate and effective way.
Conclusion : The curve and surface information extracted from CBCT image data could be used to find a normalization of the population, which is a basis for a 3-dimensional cephalometric analysis to overcome the limitations of conventional cephalometry.