Computer-assisted Orthognathic Surgery Planning Based on Profile Classification Using Self-Organizing Map

点击翻译按钮获取中文摘要

When planning surgery for jaw deformity, it is possible to generate a postoperative profile of the patient by referring to the facial profile of similar patients who have previously undergone surgery. The objective of the present study was to develop a computer-assisted orthognathic surgery planning system that searches for past cases matching the profile of a new patient, generates the postoperative profiles of the new patient, and calculates an esthetic score.Subjects were 37 patients with jaw deformities who underwent orthognathic surgery. Using preoperative and postoperative lateral cephalograms, eleven soft tissue profile measurements were taken (such as facial convexity, nasolabial angle, and esthetic line (E-line)). For each patient, these values were classified using a self-organizing map (SOM) that arranges data on a two-dimensional map based on degree of similarity. Furthermore, each cluster classified by the SOM was divided into subclusters based on movement of bony tissue caused by surgery. At this stage, an esthetic value was assigned to each subcluster based on past cases.The results of classification by the present system almost completely matched profile classifications derived from the facial morphologic classification chart. Therefore, it is possible to search for similar past cases by inputting the profile measurements of a new case. Furthermore, by referring to subclusters, it is possible to identify a surgical procedure that achieves highly esthetic outcomes. The present system is a useful tool for supporting the profile diagnosis and surgery planning of jaw deformity, counseling patients who are anxious and excited about profile changes, and obtaining informed consent.