Research Of Visualization Technology In Surgery Navigation System

Visualization in surgical navigation system is a technology which converts a series of two-dimensional medical images into a three-dimensional entity,displays the entity in the computer,and provides users interactive function according to requirement.Visualization technology is widely used in preoperative planning,medical diagnosis and medical simulation training;it is a topic of great theoretical research significance and application value.Visualization technology is studied in surgical navigation system and two algorithms are proposed.The first algorithm needs many images to construct dataset,and the second algorithm just requires an image to implement segmentation.The two methods are complementary.Firstly,patch-based segmentation using refined multi-feature for magnetic resonance prostate images is proposed,which includes three parts: the extraction of image features,the selection of extracted features and label propagation.The extracted image features and feature space are complete and comprehensive;the selected images are effective to represent image contours.Experiment shows that algorithm is efficient to segment magnetic resonance prostate images.Secondly,an automatic multi-organ segmentation of prostate magnetic resonance images using watershed and nonsubsampled contourlet transform is proposed.The algorithm combines contrast wavelet based contourlet transform and the marker-based watershed segmentation using texture gradient.Experiment shows that the segmentation results correctly correspond to the main tissues in the images,and each tissue is integrally segmented respectively with the elimination of small regions.The segmentation accuracy rate is higher than other methods under comparison.Finally,the medical image three-dimensional visualization system is completed.The main content of the system are as follows: the image is segmented into different regions;and then,the pixels in the same region are set same value,different parts of the pixels are set different values;last,these modified segmentation results are passed to reconstruction algorithm.Eventually,the system get high precision virtual human tissue model.