The Design And Implementation Of A Prototype System For The 3D Visualization Of Peripheral Nerves

The 174^th Xiangshan Science Conference held in 2001 was a sign of the start of the Chinese Digitized Virtual Human Project which was put forth by the Chinese scientists in the hope of building up the Chinese digital human with the characteristics of the East Asians which can be put into the applications in the emulation and simulation of human bodies in various fields, such as life science, clothing, aviation, space flight, car industry, and so on. The project captured the attention of the society and lots of developments have been made since then. In 2003, the 208th Xiangshan Science Conference was held and again pushed forward the research and application of the Chinese Digitized Virtual Human. Under this background, as the Academician of the Chinese Academy of Sciences in Shanghai Medical College of Fudan University, Mr. Zhongwei Chen proposed the ideas on the computer-aided 3D reconstruction of the structure of human brachial plexus. Later on Fudan University and the Institute of Computing Technology, Chinese Academy of Sciences joined together and successfully won a grant from the National Natural Science Foundation of China. The granted project "Research in the Three-Dimensional Reconstruction of the Microstructure of the Human Upper Extremity Peripheral Nerves" was then carried out. Under the support of the project, we design and implement a prototype system for the 3D visualization of peripheral nerves which is called the 3D Nerve.We place more attention on the acquirement of the topological information of the nerves in the visualization process and we hope that by doing this we can get the classification information of the nerves of different functions, so that we can look into the nerves and get much more details which can be of great help for clinical services.In this paper, we introduce the traditional basic algorithms which will be involved in the 3D visualization process, including image segmentation, contours correspondence, surface reconstruction, et al. Then we introduce 3D Nerve from the aspects of functional modules and user interface, including experimental data and environment, module of segmentation, module of contours correspondence, module of 3D visualization, module of data interface and module of interface control. Then we propose several algorithms which are essential to the visualization system, including a algorithm of multi-channel semi-automation segmentation based on Fisher Linear Discriminant, a correspondence algorithm using the moment invariants with the consideration of space restriction, algorithms of the building up of the nerves...