[Development of a computerized automatic identification system for use in cephalometry].

Objective: To establish the automatic X-ray cephalometric analysis system to simplify cephalometric steps and to provide a convenient and reliable method for cephalometric analysis.

Methods: The system which was programmed by visual-c language, and graphics and image processing techniques and artificial intelligence were used. The techniques related to computer digital image processing and pattern recognition such as Median filtering, Histogram equalization, Laplacian and Canny edge detection were introduced. It could automatically outline the contour lines of the hard and soft tissues by establishing the templates of the variable anatomical structures.

Results: The following functions were established: (1) automatically outlining the contour lines of the soft tissues. (2) automatically recognizing, measuring and analysing the landmarks of soft tissues. (3) automatically recognizing porion, sella and the landmarks of the mandible. (4) automatically building the contour lines of the hard tissues. In brief, the system used the more advanced methods, calculated more precisely and saved more time and energy than other systems.

Conclusion: The system is a more convenient and precise tool in cephalometry.