Study On The Training System Of Vascular Interventional Surgeons Based On Trajectory Prediction And Real-time Local Path Planning

With the development of virtual reality technology,the vascular interventional surgeon training system based on virtual reality technology can solve the problems of lack of realism and great difference from the living training environment.However,the risk of vascular interventional surgery is high,and the novice doctors are not skilled,so they can not accurately deliver the catheter in the operation according to the path map displayed by digital subtraction machine(DSA).In order to solve these problems,this paper designs a training system for vascular interventional surgeons based on real-time local path planning.The system can not only improve the ability of real-time correction of catheter path,but also predict the trajectory of catheter tip and early warning.Firstly,aiming at the problem that novice doctors are not skilled and have a high risk of puncturing blood vessels,this paper proposes a method to predict the trajectory of the catheter tip based on the original doctor training system in the laboratory.In this study,Kalman filter algorithm is used to realize the function of trajectory prediction.This function can early warn doctors of wrong operations in advance,and when the predicted trajectory is abnormal,the local path planning will be started in advance,ensuring the real-time performance of the local path planning.This function greatly improves the doctor's risk awareness and the safety of operation.Then,when the catheter path deviates from the DSA's path map,novice doctors lack the ability to correct the catheter path.In order to solve this problem,this paper proposes a method for dynamic local path planning of catheters based on an improved Rapidly-exploring Random Trees(RRT)algorithm.In order to effectively deal with this drawback,this paper proposes a dynamic local path planning method for catheters,which uses an improved Rapidly-exploring Random Trees(RRT)algorithm.This system can detect the predicted trajectory of the catheter tip in real time.When the predicted trajectory is abnormal,it can quickly plan out the local path,and help novice doctors to correct the trajectory of the catheter,so that the catheter can be delivered according to the planned path map.It greatly improves the ability of novice doctors to correct the catheter trajectory,and ensures the safety and success rate of the operation.Finally,the trajectory prediction experiment of catheter tip and the local path planning experiment of catheter are carried out to verify the accuracy of trajectory prediction and the training effect of local path planning.Firstly,the trajectory prediction experiment of catheter tip is carried out,and the predicted trajectory is compared with the actual trajectory.The maximum error between the predicted position and the actual position is not more than 0.5mm,and the average error is0.151 mm.Secondly,in order to verify the training effect of local path planning,a comparative experiment between no local path planning and local path planning is carried out.The experiment is divided into two parts,one part is 10 groups of experiments without local path planning,and the other part is 10 groups of experiments with local path planning.The average number of collisions in the whole process without local path planning is 10.125,and the average time-consuming of the whole process is 126.25 s.The average number of collisions in the whole process with local path planning is 1,and the average time-consuming of the whole process is99.3s.To sum up,the catheter tip trajectory prediction function and real-time local path planning function proposed in this paper improve the success rate of doctor training and greatly improve the training effect of the training system.