Medical Image Segmentation And Visualization And Their Application In The Evaluation System Of Acetabular Operation

In acetabular operations, the effect of operations is difficult to evaluate, because the 3-D structure of acetabulum is complicated and the fracture and shift of acetabulum are various. With the development of medical imaging and computer technologies, evaluation of operation assisted by computer became a hot research area.In this article, a system to evaluate the effect of acetabular operation is designed, and the segmentation and visualization parts of the system are discussed in detail. The following work is finished:(1) A segmentation method based on Canny edge detection is proposed to segment a single CT acetabular image, which is based on Canny Edge Detection, and combines with threshold, edge unification and closed boundary growth. This part provides exact data for the evaluation of operations.(2)A fast segmentation method to serial images is brought forward based on the single image segmentation, which takes advantage of the correlation between neighboring slices. This method is effective and easy to implement.(3)The volume rendering method based on Ray-Casting algorithm is proposed. The color function and lucidity function are built according to the histograms of the entire image and the object, which makes the object more distinct in the 3-D picture.Exact data ban be obtained from the segmentation and visualization parts which provides evidence for evaluation criterion of operations.Another work of this thesis is the analysis and activity prediction of Chinese medicine. A model combines Support Vector Machine and Genetic Algorithm is proposed, which predicts the activity with satisfied results. Less biology experiments and less expense are needed with this model.