Virtual Surgical Training Methods for Spinal Surgery

In this paper, we develop a virtual intervertebral surgical training system by simulating key steps extracted from real surgical workflows. The main contributions include extracting key operative steps from real surgery videos to design a complete training process, reconstructing 3D human tissue models from medical imaging data using voxel segmentation and the Marching Cubes algorithm, developing a real-time interactive soft tissue deformation model with the XPBD algorithm ,incorporating AABB and spatial hashing for collision detection and GPU acceleration, and designing system modules to simulate procedures such as bone grinding, ligament exposure, and disc clamping.