| Minimally invasive surgery is widely used in clinical medicine.Compared with the traditional bevel-tip flexible needle,the cannula flexible needle can achieve variable curvature insertion with less tissue trauma.The outer flexible cannula can isolate the friction between the flexible stylet and the tissue,and has the advantages of strong correction ability and high control precision.In order to use the cannula flexible needle for targeted insertion,this thesis studies the multi-target insertion path planning of the cannula flexible needle.According to the insertion principle of the cannula flexible needle and the non-holonomic constraint motion characteristics,a kinematic model is established.Through the theoretical analysis of the bending characteristics of the cannula flexible needle,the parameters affecting the bending deformation of the cannula flexible needle during the insertion of the simulated tissue are determined.The insertion radius of the cannula flexible needle under the different elongations of the stylet out of the cannula is measured experimentally,and the variable curvature characteristic curve of the casing flexible needle is obtained by fitting calculation.By analyzing the characteristics of the rapidly-exploring random tree(RRT)algorithm and the actual requirements of the path planning for the cannula flexible needle,the RRT algorithm based on an adaptive target bias strategy is proposed by integrating the target bias strategy and adaptive strategy of path search.MATLAB is used to simulate the improved RRT algorithm.The simulation results show that the proposed algorithm has the advantages of short running time,fast search efficiency and few iterations,which verifies the feasibility of the algorithm and can be used as the basis for subsequent path planning research.Based on the actual requirements of the insertion paths of the cannula flexible needle,the insertion path planning problem is transformed into a multi-objective optimization problem,and a mathematical model of the corresponding multi-objective optimization problem is established.The Pareto model is introduced into the artificial fish swarm algorithm(AFSA),and it is improved into a multi-objective optimization algorithm by adding a non-dominated sorting strategy and elite archiving strategy.Combining the advantages of the RRT algorithm and the AFSA algorithm,a path planning method based on the AFSA-RRT hybrid algorithm is proposed.In the generation method of random points in the RRT algorithm,the idea of foraging behavior and clustering behavior in the AFSA algorithm is introduced.The RRT algorithm is used to explore the feasible paths efficiently and quickly,and the AFSA algorithm is used to optimize the insertion paths.While ensuring the path quality,the search speed is accelerated and the solution quality is improved.The proposed algorithm is simulated by MATLAB to verify the effectiveness of the proposed hybrid algorithm.To reduce the risk of tissue damage,pain and infection caused by saturated biopsy,a multi-target insertion path planning algorithm is proposed based on the proposed path planning algorithm.Aiming at a large number of targets and a limited number of insertion entries,a target assignment strategy based on maximum and minimum distance algorithm is proposed.In order to optimize the form of multi-target insertion paths,the nearest target search strategy is proposed,and the multi-target adjacent judgment of the needle tip position is added to give a multi-target insertion sequence judgment method under adjacent or non-adjacent conditions.In order to avoid the influence of the doctor’s personal subjective judgment on the selection of the insertion entries,a needle insertion entries selection method is proposed.The effectiveness and rationality of the proposed strategy are verified by simulation experiments.In order to verify the accuracy of the variable curvature characteristics,a total of five groups of experiments were carried out.The experimental error rate was 1.33 %,which met the requirements of clinical surgery and proved the accuracy of the variable curvature characteristics.In order to verify the effectiveness of the multi-target insertion path planning method of the cannula flexible needle based on the AFSA-RRT hybrid algorithm,five groups of insertion experiments were carried out.The experimental results show that the average error is 0.3278 mm,the root mean square error is 0.5376 mm,and the maximum error is 1.0193 mm,which fully meets the accuracy of clinical surgery,proves the feasibility and rationality of the multi-target path planning algorithm,and provides a theoretical and experimental basis for controlling the cannula flexible needle for accurately targeted insertion. |
PDF Full Text Request