GPU haptic simulation of soft tissues deformation for spinal surgery training simulator

OSSim Technologies（以下简称OSSim）是一家加拿大私营技术公司， 作为基于虚拟现实（VR）的开放式骨科手术的先驱和领导者， 医学院学生和骨科住院医生。其第一款产品Sim-K模拟器可实现模拟培训， 全膝关节置换手术OSSim目前正在开发一个VR训练模拟器， 脊柱外科手术，这需要一种新的实时基于物理的模拟技术， 组织中这项研究的动机是对强大的软组织建模能力的需求日益增长， 复杂的交互式手术模拟和时间关键型应用。有限元（FE）建模，作为 精确的连续介质力学为基础的方法，将采用，以现实的模型， 脊髓、神经根和椎间盘的行为，结合最新的图示 处理单元（GPU）技术的实现，以实现高频率的更新率，这是必要的 用于触觉呈现。手术实时仿真领域，包括软组织变形、接触 建模和切割，依赖于GPU实现，以显着提高计算时间。选择 这种结合有限元建模技术和GPU计算方法的集成方案将提供 鲁棒性、稳定性和计算时间之间的最佳折衷。至于脊柱手术模拟器，触觉 设备将被纳入一个身临其境的体验与触觉，视觉和听觉反馈。掺入 在基于GPU的软组织模型内的触觉将强加额外的计算成本和挑战， 考虑到现实触觉模拟中的力反馈需要高于300 Hz的频率， 30-60 Hz用于图形。然而，这样一个实时的软组织平台将允许居民在 现实环境。OSSim将利用蒙特利尔的研究成果和 基于GPU的模型实现将允许开发高度逼真的医疗培训模拟器， 在工程、质量、制造、销售和营销方面创造高素质的工作岗位。

OSSim Technologies (hereinafter named OSSim), as a private Canadian technology company, positions itself as a pioneer and leader in Virtual Reality (VR)-based open orthopaedic surgeries designed for the training of medical students and orthopaedic residents. Its first product, Sim-K simulator enables simulation training for the total knee replacement procedure. OSSim is currently developing a VR training simulator for complex open surgeries of the spine which requires a new real-time physics-based simulation technique for deformable soft tissues. The motivation of this research is the growing need for robust soft tissues modelling capabilities in complex interactive surgical simulation and time critical applications. Finite elements (FE) modeling, as an accurate continuum mechanics based methodology, will be adopted in order to realistically model the behaviour of the spinal cord, nerve roots, and intervertebral disks combined with the most recent Graphical Process Unit (GPU) technology implementation in order to achieve high frequency update rate that is necessary for haptic rendering. Areas of surgical real-time simulation including soft tissues deformation, contact modelling and cutting, rely on GPU implementations to significantly improve computational times. The choice of this integration scheme combining FE modelling technique and GPU computational approach will offer the best trade-off between robustness, stability, and computing time. As for the spinal surgery simulator, haptic devices will be incorporated for an immersive experience with tactile, visual and audio feedback. Incorporation of haptics within GPU-based soft tissues models will impose an additional computational cost and challenge, given that the force feedback in a realistic haptic simulation requires frequencies higher than 300Hz, versus the 30-60 Hz used in graphics. However, such a real-time soft tissue platform would allow residents to train on a realistic environment. OSSim will exploit the research results in Montreal and successful outcomes from the GPU-based models implementation will allow development of highly realistic medical training simulators and creation of highly qualified jobs in engineering, quality, manufacturing, sales and marketing.