Research On Key Techniques Of Virtual Surgery Planning System Of Nasopharyngeal Carcinoma Based On CT Data Set

The modeling and simulation of human organs is one of the frontier areas of biomedicine research. It is a multi-disciplinary topic, including computer science, physics, electronics and related medical knowledge. As one of the most important and complex organs within human body, nasopharyngeal tissue's virtualization and surgery planning system plays a significant role in the practice of medicine.In this thesis, some key technologies of virtual surgery planning system of nasopharyngeal carcinoma are investigated systematically. Firstly, According to the traits of clinic CT data, we design a comparatively good filtering algorithm to filter the high-frequency interference from the CT image Data. Secondly, we investigate the segmentation result of many existing algorithms, and explore a new algorithm for extracting nasopharyngeal tissue and its surrounding pipelines. Based on the segmentation's result, we build the anatomy database of nasopharyngeal tissue. Thirdly, the 3D reconstruction technology is studied. Because of the slow rendering rate of traditional algorithms and mess medical volume date, we introduce an improved surface reconstruction method. Finally, we implement a demo system of virtual surgery planning system of nasopharyngeal carcinoma.In this thesis, some innovations had been done in the following aspects:1. According to the traits of clinic CT data, the filtering algorithm based on Gradient Information is used. Then the Canny edge detection algorithm based on the dynamic threshold is introduced in order to extract nasopharyngeal outline. The method is more efficient than the sole manual segment and more accurate than the automatic segment.2. Nasopharyngeal 3D reconstruction is implemented by the surface and the volume reconstruction methods, and the comparison is given on the advantages and disadvantages of the two reconstruction methods. Meanwhile, Because of the slow rendering rate of traditional algorithms and mess medical volume date, we introduce an improved surface reconstruction algorithm. The algorithm is proved to be simple and efficient.3. Develop a demo system of virtual surgery planning system for nasopharyngeal carcinoma diagnosis and treatment. The preparatory physical model is realized.