Research On Interactive Mechanics And Control Strategy For Robot-Assisted Percutaneous Puncture Surgery

Percutaneous puncture is a common minimally invasive surgery for the diagnosis and treatment of cancer.Robot-assisted percutaneous puncture is a significant technology for improving the surgery effect,reduce the operating burden and the damage from radiation environment to clinicians.This issue is comprehensive research integrating robotics,control engineering,biomechanics,and medical image processing.The purpose of this study is to achieve a safe and high precision puncture operation for the ideal diagnosis and treatment effect based on the processing and analysis of force,visual and other information.This work carries out in-depth research focusing on the interactive mechanics and control strategy of robot-assisted percutaneous puncture:(1)The insertion force is analyzed and modeled by considering the anisotropic mechanics of fiber-structured soft tissue.Based on the mechanics of continuum solid,the mechanics of tissue rupture during needle insertion in transverse isotropic soft tissue is analyzed.By introducing the fracture mechanics,the stiffness force of the tissue contacting deformation is modeled.According to the mechanics of unidirectional fiber-reinforced material and the transformation principle of a tensor is used to model the cutting force.The multi-angle insertion experiments are carried out on the muscle tissue,and it is verified that the proposed model can accurately describe the interaction state between the needle and the fibrous structured soft tissue.(2)Based on the insertion force models,the Euler-Bernoulli beam theory,and the Rayleigh-ritz method,the needle nonlinear deflection is modeled in the transverse isotropic tissue.Then the model is extended to the double-layered tissue driven by the interactive force respecting the needle pose.The model is verified by taking a single layer of muscle and a double layer of muscle-liver tissue as examples.The needle-accessible region in the complex interactive environment is obtained by simulation.(3)Considering the deformation mechanism of a needle,the insertion path planning algorithm is proposed by introducing the concept of cost map and artificial potential energy field.Take a liver biopsy as an example,the algorithm is simulated to obtain an insertion path in the personalized physiological anatomy environment.The personalized insertion scheme optimization algorithm for seeds implantation of current patient posture is developed.The simulation results show that the proposed algorithm can be conveniently applied to the rapid optimization of the intraoperative insertion scheme.(4)Based on the orientation-depending interactive force and deformation model,the effective operation method for the needle in anisotropic and inhomogenous tissue is determined.The control strategy optimizer is designed concerning the insertion task by using the simulated annealing method.The algorithm of needle tip automatic identification and trajectory reconstruction from ultrasound images is proposed and introduced into the needle steering equation.Then the needle intelligent control algorithm is developed based on the Model Predictive Control theory.(5)Integrating the insertion force and deflection model,the insertion scheme planning,and the control algorithm,the ultrasound-guided intelligent insertion system is developed with the control parameters identified by optimizer simulation.The insertion errors of multiple-task automatic needle insertion experiments,including static/moving target insertion and path tracking,are about 1mm which is verified that the insertion accuracy of the developed system can meet the clinical demand.