Research On 3d Interactive Visualization For Medical Data

With the advancement to the elaboration in the medical imaging and clinic research area, the demand on intuitive and comprehensive analysis of medical images is becoming urgent. Using the visualization technique, an intuitive and realistic diagnosis scene is built up in computers. So the efficiency and accuracy in the treatment is improved greatly. However, the 3D visualization model should be high-quality in the clinic application and it is time-consuming to realize. So the quality and speed are hard to account for. Some risks in the treatment relied on the difficulty in observing the 3D relationship of inner organs clearly. In addition, high configuration of graphics hardware prevents the popularity of the visualization of mass medical data in the medical area.To the questions, a quick method for 3D interactive visualization based on common computers is presented in this paper. The enhanced photorealistic method for 3D model of medical data, oblique plane positioning and extraction method, the method of dynamically separating the inner information from medical volume data and the interactive technique for mass medical data based on common computers are researched deeply. And a low-configuration, high-quality and easy-to-interacting 3D platform is brought forward for medical imaging visualization analysis. The detailed content is as following:In the aspect of improving the imaging quality, the lighting effect is introduced ultilizing the texture property of hardware to accelerate the visualization calculation. To avoid the complex computation of vertex vectors in the lighting effect, the lighting model is decomposed geometrically and composed again along the directions of the view point and lighting. Furthermore, the composed results are simulated with the help of the frame-buffer operation and the blending function of the graphics hardware. The method can guarantee not only the imaging effect but also the interactive frame rate. And the interactive visualization with a more genuine imaging quality is achieved in common computers.In order to interact with medical visualization, the visualization technique and clipping plane extracting method are united and the geometric relationship between the oblique clipping plane and the volume data is analyzed. An interactive scene for analyzing 3D data is built while the volume rendering is completed. Using the virtual interactive device and texture mapping technique, the demonstration of clipping-plane spacial relationship and dynamical clipping of 3D data is realized. Consequently, a genuine virtual diagnosis is constructed.To adjust volume data interactively, a transfer function model with a better perspective effect to separate the volume data is deduced according to the classic transfer function theory and the property of modern graphics hardware. At the same time, the interface parameters for interacting are extracted in the deducing process. By these parameters, adjusting of the 3D visualization model in real time is realized. While keeping the real-time frame rate, the different tissues hidden in medical volume data are separated according to the information of gray and gradient field. In addition, using the adjusting module, the non-linear adjusting of medical data in real time is realized to separate the hidden tissues clearly.So as to manipulate mass medical data, according to the limited throughout of common computers and the features of modern graphics, interactive volume rendering for mass data is realized in common computers using tree structure and texture mapping technique. Utilizing the octree algorithm and graphics hardware acceleration, the original volume data is divided into multi-resolution bricks, and the capacity and bandwidth restricts between graphics memory and the main memory are broken. Rendering of huge volume data with different resolution is achieved by inversing the nodes tactically. This method is helpful to lower the hardware requirement while rendering the huge medical volume data and to make it possible to interact with it with a lower configure.Finally, the content in this paper and the universal functions for analyzing the medical imaging data are united together. A platform is constructed to analyze the medical data. By that, a certain quantity of medical data can be dealt with dynamically in common computers.