Dissertation
Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device

Researches on the Key Technologies of Three-dimensional Modeling and Simulation of Virtual Eye

Author DaiPeiShan
Tutor TangJingTian;WangBoLiang
School Central South University
Course Biomedical Engineering
Keywords virtual eye eye tissue image three-dimensional modeling simulation of eye tissue deformation
CLC TP391.41
Type PhD thesis
Year 2007
Downloads 258
Quotes 1
Download Dissertation

Virtual eye tissue three-dimensional modeling is of great value to ophthalmic teaching, research, clinical diagnosis and therapy. With the development of medical imaging, medical image become an indispensable tool to eye-related disease examination and treatment. Three-dimensional models of virtual eye tissue constructed by medical images can help medical workers in multi-level observation of eye tissues, anglicizing interested area qualitatively and quantitatively. But the volume of eye tissues is very small; the structure is very complex; shape parameters are very hard to measure. Eye tissue three-dimensional modeling is very difficult.First, the key technology of virtual eye three-dimensional modeling and simulation - multi-material reconstruction technology was researched. Then, eye tissue image of CT image, retinal blood vessel image, slit-lamp micrograph and slice image were use in the eye tissue modeling method research. At last, the haptic feedback and deformation of eye tissues in ophthalmic virtual surgery were simulated.The main contents and contributions of this dissertation are as follows:(1) Level-of-detail tactic was used in the research of the key technology of virtual eye tissue three-dimensional modeling—multi-material reconstructing. Although there are many single material three-dimensional reconstruction algorithms, multi-material reconstructing algorithms have not yet mature. It is still very difficult to reconstruct multi-material three-dimensional model. Nasopharyngeal tissues reconstruction from CT series images was taken as an example of multi-material reconstructing. About dozen of nasopharyngeal tissues were reconstructed. At last, a computer simulation of nasopharyngeal carcinoma Brachytherapy method was proposed.(2) A CT image based eye tissue three-dimensional modeling method was proposed. A computer aided volumetric measurement of human orbit was proposed. Comparing with the common computer aided volumetric measurement of human orbit, the method is simple and fast, saving much energy of medical workers.(3) A fast-auto approach of cornea reconstruction based on slit-lamp micrograph was designed. Based on the symmetrical feature of corneal surface, a three-dimensional model for the 2D-image sequence registration was designed. The cornea auto reconstruction was realized. The reconstruction speed is very fast and needs no interaction during the process.(4) A retinal blood vessel image segmentation algorithm was presented. A local normalization algorithm that uniformed the local mean and variance of an image was used to eliminate background differences. Wiener filter was used to smooth the image, Expectation-maximization algorithm was use to classify the pixels into several class, and noise removing step was used to remove small speckles in the segmented image. Compared with some existing algorithms, the algorithm can segment the blood vessel automatically, need not manual interaction. The input parameters are comparatively fixed. The segmentation results are good. Based on the anatomical and physiological characteristics of the eye, a reconstruction method to restore three-dimensional mode from retinal blood vessel image was realize by inverse projection theory. The three-dimensional model is close to the real distribution of retinal vascular in eye fundus. The model can be observed from different direction.(5) A geometry model constructing method based on eye tissue slice image was brought forward. Geometry model of cornea, sclera, iris, lens, vitreous and other organizations were constructed, maintaining mutual position relationship. The three-dimensional models constructed here are better than the models commonly used by finite element analysis. The method can also be use to modeling rotating body with complex surface. Based on the eye tissue models, the tissue deformation and haptic feedback of eyeball in eye surgery were simulated. The haptic feedback of the eyeball consists of the visco-elastic force and sticking-sliding friction force. The deformation of eyeball was simulated by mass spring model.

Related Dissertations
More Dissertations