Feature preserving refinement of surfaces for web-based surgical simulation.

Abstract

In plastic surgery, 3D models of the affected part are often used for the purpose of visualizations and surgical simulations. The optimal models for web-based surgical simulations keep high accuracy in affected parts and keep low accuracy in other parts. Consequently, the data size becomes small. In this research, we propose a method to generate free-form surfaces based on Lattice Structure from polygonal meshes. The polygonal meshes are generated automatically from CT and MRI data using Marching Cubes. By changing the resolution of input images, the accuracy of output meshes is controlled. Free-form surfaces based on Lattice Structure are fitted to polygonal meshes. Lattice Structure is a method to manage a free-form surface with a simple base polygonal mesh. The data size is quite small because surface shape is converted and saved as a simple polygon. A free-form surface is quickly generated and high accuracy is maintained. Moreover, users can input character lines and they are reflected as boundaries of patches. The models generated with this method are partly accurate and compact. These data make it possible to simulate surgery on the WWW, because they can be quickly transferred