The Clamping Prediction And Simulation Of Removal Loose Body In Virtual Knee Arthroscopic Surgery

In the treatment of knee injuries, removal loose body of the knee is more common. It’s one of the most suitable arthroscopic treatment of knee surgery. But the arthroscopic knee surgery operation difficulty, lack of training tools and long cycle, so that its popularity is limited. The traditional arthroscopic surgery training has a high cost, low recycling rate and other shortcomings, due to the virtual surgery system with visual simulation, human-computer interaction and force feedback, which alleviate the problem of physician training to some extent by the simulation system.For arthroscopic of removal loose bodies of the knee, the most critical step is to clamp and remove loose bodies in vitro. In order to improve the virtual surgery system which the fidelity of visual simulation, behavior model, and real-time and accuracy of the contradictory issues, this dissertation take arthroscopic removal loose body of the knee as the research object, for the study of creating a visual and collision model to the virtual arthroscopic knee surgery system, real-time collision detection, clamping prediction, and other key technologies. The main research contents of this dissertation are as follows:(1) Reconstruction three-dimensional model of the human knee. As used herein, the medical data of human knee MRI image sequences, image preprocessing, image segmentation and three-dimensional reconstruction, and rebuilding the model surface smoothing treatment and meshing. The ellipse fitting is used in the image of loose bodies, and then creating an ellipsoid with visual and collision model that can satisfy the simulation of loose body removal surgery. These models are more realistic to show the reality and improve the immersion in virtual surgery system.(2) Collision analysis. By assuming that the loose body is a rigid body can be translated into the ellipsoid, virtual knee loose body and blue clamp of surgical instruments for clamping motion analysis and collision analysis. Based on points, lines, polygons, and AABB bounding box hybrid ray tracing collision detection algorithm was used, and set up octree that accelerate the establishment of bounding box algorithm, improve the timeliness and accuracy of the virtual surgery of removal loose bodies in the clamping collision detection.(3) Clamping prediction. According to the clamping collision detection algorithm to judge the success of clamping the loose body, in a stationary state during the operation that making stress analysis of the blue clamp clamping loose body, and loose body and blue clamp collision response to feedback information of the collision force and collision point, to design a clamping prediction algorithms to achieve the success of clamping loose body prediction, improve operation training feasibility of doctors in the virtual knee arthroscopic of removal loose bodies.(4) The simulation. Based on the above research and development of a prototype experimental system, this dissertation use SOFA, the medical simulation software, development of a virtual arthroscopic system of removal loose bodies of the knee with force feedback function, on the loose body clamping of the collision detection and to clamping prediction test. Surgical simulation system process is operable, and can realize the simulation of the free body clamping, which can meet the basic needs of removal loose bodies.