Experimental Study Of Soft Tissue Deformation Modeling Method

In the traditional surgery training, people often use animal carcasses as surgery training object, but because of the body and rich life living tissue there is biggish difference in the aspect of mechanical properties, leading to training effect is not satisfied. With the rapid development of computer technology, scientists will be attention to virtual surgery simulation system, and the technology of the research in many aspects. By building more perception of virtual surgery system, can provide doctors with surgical plan formulation, preview, the results of prediction and computer assisted surgery, and other functions. Virtual surgery system is the important application on the virtual reality technology in modern medicine, the birth and development of this technology for doctors to provide a lot of convenience, at the same time also makes the general patients benefit from it. At the same time, the current virtual surgery system of truth and the real scene there is large gap, how to build a more complete virtual surgery system is still a challenging task.Paper analyzes the current research of soft tissue modeling method, in view of the existing modeling of virtual surgery,usually adopt theoretical derivation is modeled with the true model gap defects, as well as the traditional modeling method focuses on the visual effect of soft tissue deformation, and the lack of analysis of mechanics effect,resulting in the deformation model derived from the theory and reality of soft tissues under stress behavior exist a certain differences. Mass-spring and therefore, in this paper, the least squares support vector machine (SVM) modeling method to compare the two kinds of soft tissue deformation,put forward and build the data acquisition system, and through the data acquisition system to obtain accurate stress, strain data of soft tissue, and then use based on support vector machine(SVM) to carry on the system identification, the experimental data of soft tissue deformation model is established. The main work is as follows:Firstly,the soft tissue deformation model is established by using mass spring model.Through in-depth analysis and comparison of finite element model and the traditional mass -spring modeling method, this paper adopts the mass spring model as the first research method to establish a model of soft tissue deformation; application of 3D Studio MAX constructed a geometric model of the soft tissue, and visual rendering using OpenGL 3D graphics library.Secondly, the method of collecting soft tissue deformation data by data acquisition system is put forward. For data collection are convenient to be soft tissue module carries on the grid division; for the acquisition to the accurate deformation data, constitutes a data acquisition system using a keen scholar LK-G155 laser rangefinder, ATI DAQ MEIA force acquisition device and control device, combined with image processing method of acquisition and calculated according to the pressure of horizontal deformation and vertical deformation displacement deformation data.Again, we based on least squares support vector machine (SVM) modeling of soft tissue.By analyzing the requirements of this article, the design identification experiment steps;Selected generalization ability strong least squares support vector machine (SVM) as the system identification method; Detailed modeling steps; Build out of the soft tissue deformation modeling identification system.Finally, the application of least squares support vector machine (SVM) method to establish the model and the real model for comparison, according to the mean variance to judge by using the least squares support vector machine (SVM) model is reasonable; of the will use the mass spring method to build model and application of least squares support vector machine (SVM) model were compared, the analysis shows that by using the least squares support vector machines identification model is more consistent with the relation of stress and strain stress in soft tissue.