摘要
目的:尝试将数字孪生(DT)技术应用于人体骨骼分析,探索人体骨骼DT体开发的框架及其关键技术。方法:针对人体腰椎骨骼,利用虚拟现实(VR)动作捕捉技术和传感器技术采集人体重要位置处的数据,将所得大量实时传感器数据通过人工智能技术进行数据分类、筛选、约简与计算处理得到所需要的关键数据。通过人体反向动力学与生物力学算法对关键数据求解得到目标骨骼的空间方位信息与力学信息,将部分传感器数据与计算结果融合后对目标骨骼进行仿真模拟,得到目标骨骼的各种生物力学性能,并利用代理模型技术对未知姿态下骨骼的生物力学性能进行预测,最后利用计算机图形学技术将所得性能数据建模渲染得到真实骨骼的高保真DT体,实现对腰椎骨骼生物力学性能的忠实孪生映射。结果:人体DT构建系统包括物理装备系统、数据传输系统、数据处理系统、数值计算系统、数据融合系统、仿真预测系统、孪生显示系统及数据存储系统。通过各系统协同作用,利用穿戴式VR设备和少量传感器即可实时计算出目标骨骼的生物力学性能,结合实验数据对骨骼生物力学性能的评估、预测及反馈控制,实现对目标骨骼实时的健康监测。结论:基于DT技术可实现人体骨骼实时动态的高保真数字化显示和运动动作的实时模拟,同时可提供骨骼的生物力学性能动态仿真分析、损伤预测和危险姿态预警等,为人体骨骼疾病防治及骨骼健康的监测提供理论依据和方法参考。
Objective: To apply digital twins(DT) technology to human bone analysis, and explore the framework and key technologies of human bone digital twins development. Methods: For real human lumbar spine bones, virtual reality(VR) motion capture technology and sensor technology are used to collect data at important positions of the human body, and the large amount of real-time sensor data obtained is classified, filtered, reduced and calculated by artificial intelligence technology to obtain the key data needed. Key data is solved through human inverse dynamics and biomechanics algorithms to obtain the spatial orientation information and mechanical information of the target bone. After fusing part of the sensor data with the calculation results, simulate the target bone to obtain various biomechanical properties of the target bone. And use proxy model technology to predict the biomechanical properties of the bones in unknown poses;finally use computer graphics technology to model and render the obtained performance data to obtain a high-fidelity digital twin of the real bones, to achieve a faithful twin of the biomechanical properties of the lumbar spine bones mapping. Results: Human digital twins construction system includes physical equipment system, data transmission system, data processing system, numerical calculation system, data fusion system, simulation prediction system, twins display system and data storage system. Through the synergy of various systems, the biomechanical properties of the target bone can be calculated in real time using wearable VR devices and a small number of sensors under a variety of human postures, combined with actual experimental data to evaluate, predict and feedback control the biomechanical properties of human lumbar spine bones, and to achieve real-time health detection of target bones to prevent lumbar spine injury. Conclusions: Based on the digital twins technology, it can realize the real-time dynamic high fidelity digital display of human bones and the real-time simulation of sports movements. At the same time, it can provide solutions such as dynamic simulation analysis of bone biomechanical properties,damage prediction and dangerous posture early warning. This provides a theoretical basis and method reference for the prevention and treatment of bone diseases and the monitoring of bone health in the future.
作者
李忠海
孙天泽
张警
何西旺
宋学官
LI Zhonghai;SUN Tianze;ZHANG Jing;HE Xiwang;SONG Xueguan(Department of Orthopaedics,The First Affiliated Hospital of Dalian Medical University,Dalian 116011,Liaoning;School of Mechanical Engineering,Dalian University of Technology,Dalian 116024,Liaoning,China)
出处
《中华骨与关节外科杂志》
2022年第2期111-116,共6页
Chinese Journal of Bone and Joint Surgery
基金
辽宁省“兴辽英才计划”项目资助(XLYC1807131)。
关键词
数字孪生
数字化医疗
人体反向动力学
骨骼
机理模型
Digital Twins
Digital Medicine
Reverse Human Body Dynamics
Bone
Mechanism Model
