Three dimensional CT scans and facial light scans provide rich new sources of data for the study of craniofacial anomalies. Thus far these sources have principally been used to illustrate details of pathologic anatomy on individual cases and have been the data source for surgical simulation software used for treatment planning. What is necessary for these modalities to realize their potential is a method for the analysis of surface form. Most of the craniofacial skeleton is composed of long stretches of smoothly curving surfaces punctuated by prominent ridge curves and occasional point landmarks. Landmark based analyses are extremely well developed, but are not applicable to the surfaces between them. This subproject intends to develop an analysis method which can be used to study these surface regions. Differential geometry and its discrete computational geometry counterpart are the principal mathematical tools which will be used to analyze and describe surface form. The principal curvatures at any point on a surface can be extracted using standard computational methods. Using these numbers all of the principal ridges on a smoothly curving surface will be extracted semiautomatically. These ridge curves will be the first geometric features to be analyzed statistically in our craniofacial patient subpopulations. Future work will use these ridge curves to bound surface regions for more comprehensive analysis.
Showing the most recent 10 out of 13 publications
