基于非接触组织振动测量的骨外科手术机器人钻磨状态精准辨识

Robotic systems in orthopaedic surgery are widely used to perform bone drilling and milling operation. In order to prevent possible damages to the normal tissues caused by high-speed rotating drill and burr, identification methods are required to provide valuable information on the condition of the bone and its adjacent soft tissues during surgery. The accurate identification method we used is realized via the measurement of tissue vibration amplitude during cutting process, which is directly recorded by a non-contact laser displacement sensor. The main research contents of the project include three aspects. First, during drilling process the vibration signal of bone is recorded by the laser displacement sensor, and then wavelet packet transform is performed to extract the energy spectrum from the measured data, so we can obtain the relation model of the penetration depth of the drill and the wavelet energy spectrum, and the penetration depth can be estimated in real time. Second, during milling process the vibration signal of the bone being cut, as well as its adjacent bones, is recorded by the laser displacement sensor, and variation of wavelet energy spectrum is used to detect the boundary of bone after spike noise is detected and replaced. Third, during milling process the vibration signal of the bone and its adjacent soft tissues is recorded by the laser displacement sensor, and then the wavelet energy spectrums related to different tissues are compared to discriminate tissue types. The proposed method is experimentally verified through the drilling and milling operation in porcine spines.

骨外科手术机器人系统常用于完成骨组织的钻孔和磨削操作，为了防止高速旋转的钻头和磨头对正常组织的损伤，在手术过程中需要对骨组织及其周围软组织的状态进行辨识。本项目拟利用非接触激光位移传感器直接测量组织的振动实现对钻磨状态的精准辨识。主要研究内容拟分为以下三个方面：首先由激光位移传感器采集钻孔过程中的骨振动信号，利用小波包变换获得信号的能量谱，通过建立小波能量谱与穿透深度的映射关系模型实现骨组织穿透深度的实时估计。其次，由激光位移传感器分别采集磨削过程中被切削的骨骼和相邻的骨骼的振动信号，在尖脉冲噪声被检测和替换之后根据小波能量谱的变化来识别骨骼的边界。最后，由激光位移传感器采集磨削过程中的骨组织及其周围软组织的振动信号，从而能够通过比较小波能量谱的差异来达到识别组织类别的目的。本项目拟通过在猪脊柱上进行的手术验证提出的钻磨状态辨识方法的有效性。

骨外科手术机器人系统常用于完成骨组织的钻孔和磨削操作，为了防止高速旋转的钻头和磨头对正常组织的损伤，在手术过程中需要对骨组织及其周围软组织的状态进行辨识。本项目利用非接触激光位移传感器直接测量组织的振动并利用麦克风测量切削声音实现对钻磨状态的精准辨识。主要研究内容包括以下三个方面：首先，由激光位移传感器分别采集磨削过程中被切削的骨骼和相邻的骨骼的振动信号，在尖脉冲噪声被检测和替换之后根据信号幅度的变化来识别骨骼的边界。其次，由激光位移传感器采集磨削过程中的骨组织及其周围软组织的振动信号，从而能够通过比较小波能量谱的差异来达到识别组织类别的目的。最后，考虑声信号能提供关于切削刀具及组织接触状态信息，通过采集、分析在椎板切除术中产生的声信号实现了切削状态监测。通过在猪脊柱上进行的切削手术验证了提出的钻磨状态辨识方法的有效性。