Research On Image Registration Method In Surgical Navigation And Robot System

With the continuous implementation of the national "precision medicine" strategy,people have put forward higher and higher requirements for medical quality.As an important tool for precision medicine,surgical navigation and robot system can effectively help doctors to make preoperative planning and assist doctors to complete the surgery through surgical navigation.Image registration is a key technology in surgical navigation and robot systems.Its registration accuracy and speed directly affect the system's auxiliary effect on surgical operations.The iterative closest point algorithm is one of the simpler and better effects in the 3D image registration algorithm,but the algorithm will be affected by the size of the two sets of paired point sets when looking for the closest point pair set.This requires the introduction of intelligent optimization algorithms such as differential evolution algorithm to help the iterative nearest point algorithm to find the nearest point pair set,thereby improving the image registration effect.This paper firstly improves the differential evolution algorithm from the perspective of the optimization algorithm.By comprehensively using the search space and fitness space information,this paper proposes a differential evolution algorithm based on the comprehensive ranking of distance and fitness.This algorithm selects some individuals through comprehensive inter-individual distance and The fitness ranking information is obtained.This method only changes the mutation step of the classic differential evolution algorithm and can be directly integrated into the commonly used mutation strategy.Secondly,the improved optimization algorithm is added to the iterative closest point algorithm,using the comprehensive ranking based on distance and fitness.The differential evolution algorithm improves the efficiency of finding the nearest point pair set,and thus improves the speed and accuracy of image registration.The experimental results show that the ratio of neighboring groups in the improved differential evolution algorithm is related to the number of disturbances in the mutation strategy.The improved algorithm under the optimal neighboring group ratio has better computing power than the existing algorithms,and the more complex the problem,the more obvious the advantage;After determining the mutation strategy and its corresponding optimal neighboring group ratio,the improved iterative nearest point algorithm reduces the running time on the basis of ensuring accuracy compared with the classical iterative nearest point algorithm.This alleviates the bottleneck of the iterative closest point algorithm running slowly due to the large point set size.