| Virtual surgery is applied in the field of modern medicine more often.It combines the advantages of computer science and modern medicine,and it makes a significant contribution to the development of modern medicine.The appearance of blood flow is very common in the surgeries.In fact,the blood flow is often related to the success or failure of the operation.Therefore,it is necessary to do the research on the blood flow simulation in the virtual surgery.Because of the problems of low accuracy,poor real-time and poor interaction sensitivity in the traditional blood flow simulation methods,it is necessary to improve the real-time and real-time of blood flow simulation and to realize the real-time simulation of blood flow in virtual surgery,this paper combines the method of FLIP and the relevant theoretical algorithms of fluid-solid interaction based on data-driven method.The main work of this paper are as follows:(1)Basic work for blood flow data acquisition,analysis and modeling.Based on the Navier-Stokes(N-S)equations,then analyzes the basic process of solving the input data,combined with the specific application point of artery bifurcation,this paper analyzes the relevant factors that affect blood flow here.Finally,it establishes the mapping between input data and related factors,also called modeling,which lays a theoretical foundation for the later research work.(2)Data-driven blood flow simulation.Combining FLIP(Fluid-Implicit-Particle)method to solve N-S equations,the input data are obtained.Using the input data to solve the velocity field of the blood flow surface.The solution is divided into three steps: discrete solution,iterative calculation and surface projection.In the process of iterative calculation,there is a problem that the accuracy of the solution is low.In this paper,multi-scale iteration and residual iteration are proposed to optimize the results of the solution to improve the details of the blood flow surface.Finally,the Screen Space Fluid(SSF)method is used to draw the blood flow surface,then the data driven blood flow simulation is completed.(3)Blood flow simulation in interactive scenes.In this paper,real-time simulation of the interaction between scalpel and blood flow is carried out by introducing the scene of interaction between scalpel and blood flow on the basis of data-driven.Firstly,the particle solution of scalpel and blood flow is used to solve the problem of insufficient interaction due to the large boundary particle spacing in fluid-solid interaction.In this paper,the Poisson sampling method is improved.And the final sample particles were selected by twice Poisson conditions,then combined with the grid recursive method proposed in this paper,the computational efficiency of particle attribute solving process is improved to improve the real-time performance of interactive simulation.Finally,the real-time simulation of the interactive process is completed by reconstruction and real-time rendering of the interactive surface.(4)Build a prototype system of blood flow simulation based on data-driven.The work in(1),(2),(3)is modularized,and different functional modules are constructed to realize the corresponding blood flow simulation.The corresponding simulation results and brief analysis are given.The simulation results show that the proposed data-driven blood flow simulation methods show the high accuracy and high efficiency,and the simulation is realistic and real-time,which is suitable for blood flow simulation in related scenes of virtual surgery. |
