Research On Key Technologies Of MR Image-Guided Robot-Assisted Percutaneous Surgery

Magnetic Resonance(MR)image-guided robot-assisted percutaneous surgery is the most common method used in the clinical tissue biopsy or brachytherapy seed implantation.This system is a comprehensive research integrating the discipline of medical image processing,robotics,biomechanics,which contributes to achieve surgery with high accuracy and high reliability,and then improve the success rate of tissue biopsy or the cure rate of patients.This dissertation is aiming to improve the accuracy of the MR image-guided robot-assisted percutaneous surgery,and the key technologies are deelply studied.The main contents are shown as follows:(1)A standardized image quality analysis system is developed,in which the relative change of the signal to noise(SNR)and image change factor(ICF)between two series images are calculated.The standardized MR scanning sequence is obtained through the computing of the relative changes and comparision of the different scanning sequences.The effects of the robot and robot motion on the image quality are also analyzed through the comparision of the relative change of SNR and ICF.A model-toimage registration method with high accuracy is developed,which also helps measure the quantitative image distortion and improves the accuaracy of the MRI image-guided robot-assisted percutaneous surgery using rigid needles.(2)An interactive force model between the surgical needles and multi-layer soft tissue is developed to solve the challenge of predicting the forces between the needle and the real tissues in clinical trial.In the force model,the stiffness force is constructed based on the Sneddon integral method and the friction force is derived based on the modified Winkler foundation model.The impact factors are mainly calassified as the needle geometric parameters,soft tissue characteristics and the insertion methods.Then the effects of the impact factors on the insertion forces are investigated,which drives a guide of the future research on the deflection analysis between the needle and soft tissues.(3)Through the nonlinear Winkler foundation model and the Euler-Bernoulli beam theory,the needle deflection model is developed.A trajectory planning algorithm of the flexible needle insertion into soft tissue is developed based on the artificial potential field method,which intergrates the predicted deflection of the flexible needle and make it serve as a constrain of the search method.The the planned trajectory in the anatomical environment is obtained.(4)A modified 3D kinematic model of the flexible needle is developed based on the Webster bicycle model and the duty cycle algorithm.Through the change of the duty cycle factor,the needle can achieve paths with different curvatures.The empirical relationship between the duty cycle factor and the needle curvature is obtained through a series of expreiments.Heading error and cross track error controllers are construted to eliminate the heading error and the corss track error of the flexible needle in real time,which solves the problem of the flexible needle steering and path tracking.(5)A pelvis organ group mockup is designed and manufactured,which is utilized to mimick human tissues and solve the gap that the tissue mimicking is focused on a speficied single organ.Different types of the PVA hydrogels with different concentrations are prepared.The microstructure characteristics and deformation properties of the PVA hydrogels and the human tissue are observed and compared.Through a series of processes including the image segmentation,reconstruction,and anatomical research,the organ group mockup is manufactured.Through the insertion force test,it is proved that the organ group mockup has similar insertion force with the real tissues.Through the needle steering test,the accuaracy of the robot-assisted percutaneous surgery using flexible needles is approved to be 1.15 mm,which meets the surgical standard.