Medical 3D Visualization Based On Volume Rendering

Medical image 3D rendering not only can improve the level of medical diagnosis, but also play an important role on surgery planning, anatomy education and medical research. Surface rendering and volume rendering are two kinds of displaying techniques in Medical Visualization. Furthermore volume rendering is a whole new method compared with the traditional rendering techniques. This paper discusses the Medical Visualization based on volume rendering, which is a hot topic in related fields.In this thesis, visualization technique is applied for medical image data analysis and processing. Taking into account the need for medical image processing and the current condition of medical equipment in china and the characteristic of the CT or MRI image data, we make use of new visualization technique and achievement to process the medical image data in computers. In this way, many functions that can't be achieved in normal equipments can be achieved. Thus the processing functions can be more, the information can be richer, the operation can be more convenient. The goal of my research mainly optimized the algorithms of 3D rendering and integrate some accelerating algorithms of volume rendering to achieve high rendering speed and finish a 3D rendering software packages based on the personal machine (PC).The work and achievements completed in this paper consists of:Firstly, this paper systematically investigates some key issues of volume rendering. The related work includes algorithm principle, key techniques and accelerated techniques. Secondly, we analyses the volume rendering based on ray casting, includeing algorithm principle, implementation pipeline .And for the features of medical slice images, improved trilinear interpolation algorithm is presented with an increase of the speed ray-casting in direct volume rendering.Thirdly, In order to achieve higher rendering speed, we adopt the shear-warp algorithm. The shear-warp algorithm runs faster than other methods at present. It operates by factoring a projection to form an intermediate.Therefor it can run better.This paper deeply investigates the shear-warp algorithm.But two factors caused the distortion of the final image in this algorithm. This paper presented a method to correct them. This method got the final image by computing two different groups of resample points and projecting them. The first group of resample points is computed in the way just like the ray-casting method. And the second are gotten like the ones getting by shear-warp algorithm, so the new method can run better and rather fast. Examples showed that it is an affective method.