Development of a virtual reality-based zygomatic implant surgery training system with global collision detection and optimized finite element method model

: Zygomatic implant surgery is challenging due to the complex structure of the zygomatic bone, limited visual range during surgery, and lengthy implant path. Moreover, traditional training methods are costly, and experimental subjects are scarce. : To overcome these challenges, we propose a novel training system that integrates visual, haptic, and auditory feedback to create a more immersive surgical experience. The system uses dynamic bounding volume hierarchy (BVH) and Symplectic Euler to detect global collisions between surgical tools and models, while an optimized finite element method (FEM) model simulates soft tissue and detects collisions. Compared to previous works, our system achieves global rigid-body collisions between surgical tools and patient models, while also providing stable and realistic simulation and collisions of soft tissues. This advancement offers a more realistic simulation for zygomatic implant surgery. : We conducted three experiments and evaluations. The first experiment measured the axial force generated during the zygomatic implant simulation process and compared it with actual surgery, demonstrating the realistic force rendering feedback of our system. The second evaluation involved 15 novice surgeons who experienced the system and completed a questionnaire survey focusing on five aspects. The results showed satisfactory evaluations. The third experiment involved six surgeons who underwent in-depth training for two hours daily and were tested on the first, third, and fifth days. We collected data and combined it with the doctors' feedback to prove that our system can improve surgeons' proficiency in zygomatic implant surgery and provide a novel training solution for this procedure. : We have innovatively integrated global collision detection and optimized soft tissue simulation into our system. Furthermore, we have conducted experimental validation to demonstrate the effectiveness of this implementation. We present a novel solution for zygomatic implant surgery training.