Quality Evaluation And Optimization Of Robot Assisted Sutured Vessel Anastomosis

The robot-assisted anastomosis holds great promise for the future in many surgical procedures involving blood vessels. Compared with traditional surgery, high precision and accuracy can be obtained in robot-assisted surgery, especially in microvascular reconstruction. A robotic system can provide quantitative information of tissue dimensions, supply force feedback, and enhance the surgeon's vision and kinematic capabilities. However, robot-assisted surgery also has some limitations, i.e., the lack of haptic feedback and high eye-hand coordination required of the surgeon. Because of the difference in the operations between traditional surgery and robot assisted surgery, there is a new challenge in ensuring the sucessful completion and quality of surgical procedures. This dissertation develops new quality evaluation criteria based on the physical process and clinical requirements of end-to-end vessel anastomosis. Mathematical models are developed for the optimization of the end-to-end vessel anastomosis. The relationships between surgery quality and process factors are obtained using statistical methods. The contributions of the disseration are as follows:(1). A new method is developed to obtain the allowable limit of the suture tension in order to avoid blood osmosis and keep the tissue free from injury. Based on the classical surgery evaluation system, the new quality evaluation criteria and process factors are presented for the robot-assisted anastomosis. A three-dimensional finite element model of anastomosis is presented to establish the mechanical relationship between the vessel and sutures. The stress distribution of the vessel loaded by the suture is calculated using finite-element simulations and the limit of the suture tension is given to allow successful surgical tasks.(2). A mathematical model, including optimization variables, multi-objective functions and constraint conditions, is established to describe the optimization problem in robot-assisted vessel anastomosis. Simulation experiments are arranged by design of experiment to obtain the allowable tension range of suture and distribution of tissue stress based on finite element model of surgery process. The relationship between the objective functions and process variables is extracted from experimental data. A Pareto optimal solution is presented which aids surgeons in process parameter selection according to applications. (3). A robust design is carried out to consider the influence of the noise variables on the relationship between surgery quality and design variables. The noise variables investigated include the wall thickness of blood vessel and vessel property described using material constitutive equation. Based on such robust design, the effects of design variables on the mean value and standard deviation of the objective function are analyzed, and the influence of noise variables is also discussed. Furthermore, a Pareto optimal solution is also provided with the objective of increasing the S/N ratio, which is defined as the ratio of mean response over standard deviation. Finally, results from traditional optimal design are compared with those obtained by robust design.