面向机器人手术的脊柱运动建模与补偿研究

发布时间：2018-04-09 18:57

本文选题：术中脊柱运动建模　切入点：噪声处理　出处：《哈尔滨工业大学》2017年硕士论文

【摘要】：椎弓根内固定术是最常见的脊柱手术术式,由于脊柱特殊的解剖结构,椎弓根内固定术也是公认的风险最高、难度最大的外科手术之一。以此为背景,越来越多的研究着力推动辅助手术机器人在椎弓根钉内固定术中的应用。但是由于在手术过程中生理运动会引起脊柱位置准周期的微小起伏,兼之椎弓根安全区域狭窄、手术机器人刚度较大,使得脊柱辅助手术机器人系统在辅助医生钻钉道过程中的安全性难以保证,阻碍了机器人手术系统在临床上的应用。本文在研究术中脊柱起伏原因的基础上,提出基于潮气量实时预测脊柱位置偏差的模型,并设计合理的主动补偿方案。本文在参照传统手术流程的基础上,对机器人辅助手术的流程进行规划,并量化分析了脊柱起伏对手术安全的影响。为了获得生理信号数据,基于Arduino UNO开发板、红外跟踪设备等搭建生理信号采集平台,主要分为脊柱位置采集模块和生理呼吸信号采集模块,并分析在不同呼吸模式下(深呼吸模式与平静呼吸模式)采集得到的数据特点。由于采集的过程中必然存在干扰,为了获得更真实的信号,分别利用快速傅里叶变换与小波变换对脊柱在空间三个方向的运动信号进行处理。针对小波分析中小波基函数难以选择的问题,我们引入均方差决策法,根据实际滤波效果挑选出最合适的小波基。由于直接导致脊柱位置变化的原因是胸腔的扩张与收缩,本文参考胸腔结构建立双椭圆模型,并以其交点代表脊柱位置。计算得到交点表达式后,利用粒子群优化算法进行参数辨识,拟合的效果由决定系数(R2)定量给出。在研究3-PRS并联平台逆解的基础上,用其模拟术中脊柱生理运动,并将新鲜猪板骨固定在并联平台上。同时使用脊柱手术机器人在猪板骨上进行钻孔实验,通过对比实验分别获得理想情况下(骨头不动)、没有补偿情况下(骨头动,手术机器人未补偿该运动)与机器人主动补偿情况下(骨头动,手术机器人补偿该运动)三种不同的钻削力。通过对钻削力进行单因素差方分析,得出机器人主动补偿方案可以很好的消除脊柱生理运动对机器人辅助手术的影响。
[Abstract]:Pedicle internal fixation is the most common spinal surgery. Because of the special anatomical structure of the spine, pedicle internal fixation is also recognized as one of the most risky and most difficult surgical procedures.Against this background, more and more researches are focused on the application of assisted surgery robot in pedicle screw fixation.However, due to the slight fluctuation of quasi-cycle of spine position caused by physiological exercise during surgery, and the narrow safety area of pedicle, the rigidity of surgical robot is larger.It is difficult to guarantee the safety of spine assisted surgery robot system in the process of assisting doctors to drill the nail, which hinders the application of robot surgery system in clinic.Based on the study of the causes of intraoperative spine fluctuation, a real time prediction model of spine position deviation based on tidal volume is proposed, and a reasonable active compensation scheme is designed.Based on the traditional operation procedure, this paper plans the procedure of robot assisted surgery, and quantitatively analyzes the influence of spinal undulation on the operation safety.In order to obtain physiological signal data, a physiological signal acquisition platform is built based on Arduino UNO development board and infrared tracking equipment, which is mainly divided into spinal position acquisition module and physiological respiratory signal acquisition module.The characteristics of data collected in different breathing modes (deep breathing mode and calm breathing mode) were analyzed.In order to obtain more real signals, the fast Fourier transform (FFT) and wavelet transform (WT) are used to process the motion signals of spine in three directions.In order to solve the problem that wavelet basis function is difficult to select in wavelet analysis, we introduce mean-square decision method to select the most suitable wavelet basis according to the actual filtering effect.Due to the dilation and contraction of the thoracic cavity, a double elliptical model is established with reference to the structure of the thoracic cavity, and the position of the spine is represented by the intersection of the two elliptical models.After the intersection expression is obtained, the particle swarm optimization algorithm is used to identify the parameters, and the effect of fitting is quantitatively given by the determinant coefficient (R2).On the basis of studying the inverse solution of 3-PRS parallel platform, this paper simulates the physiological movement of the spine during operation, and immobilizes the fresh porcine plate bone on the parallel platform.At the same time, the spinal surgery robot was used to drill holes on the pig plate bone, and the ideal condition (bone immobility, no compensation) was obtained by contrast experiment.There are three different drilling forces between the operation robot without compensating the motion and the robot with active compensation (bone movement and surgical robot compensation for the motion).Through the single factor differential analysis of drilling force, it is concluded that the robot active compensation scheme can eliminate the effect of spinal physiological motion on robot assisted surgery.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：R687.3;TP242

【相似文献】