Knowledge-based cephalometric analysis: a comparison with clinicians using interactive computer methods.

Abstract

In modern orthodontic practice great reliance is placed on systematic and objective methods of characterizing craniofacial forms, using measurements based on both hard and soft tissue landmarks. Lateral skull X-ray images are routinely used in cephalometric analysis to provide quantitative measurements useful to clinical orthodontists. It is argued that a model- and knowledge-based methodology provides the best approach in successfully interpreting digitized lateral skull radiographs. A rule-based segmentation system, making use of an image appearance model, is used to extract image features from gray-level images. Complex image features and cephalometric landmarks are constructed from these segmented component features. A predictive model, defining picture structure, allows location hypotheses to be made for image features. The underlaying structure of the location model provides the basis for a geometric constraint model of use in discriminating between image feature candidates. A blackboard system is used to organize these tasks hierarchically, with individual knowledge sources grouped according to function and the individual stages of the adopted image interpretation cycle. Quantitative results demonstrate the superiority of this complex system over its component segmentation system run on its own. Comparisons with clinicians demonstrate both the strengths and the weaknesses of the present system. Comparisons with previous systems are favorable.