| As a minimally invasive operation,puncture intervention plays an important role in clinical practice with the characteristics of less trauma and fast recovery effect,but there are also a series of problems that need to be solved.When planning the puncture path,doctors usually use ultrasound to obtain the image field of view,but the image space obtained is limited,the quality is insufficient and the depth information is lacking,which makes it difficult for doctors to judge the information of the surgical area,and the judgment process is also extremely dependent on the doctor’s personal experience and imagination,which reduces the surgical quality and efficiency of puncture intervention.In view of the above problems,this thesis applies the augmented reality system and path planning algorithm to the flexible needle puncture path planning,and studies the prostate puncture trajectory planning based on augmented reality.This method solves the insufficient image information,obtains good medical three-dimensional images,and combines the obtained three-dimensional image information with the improved path planning algorithm to plan the puncture path of flexible needles.The prostate magnetic resonance(MRI)image was preprocessed,and the fuzzy C-means(FCM)clustering and segmentation algorithm with improved ant colony algorithm was used to segment the preprocessed image and extract the characteristic information of the prostate gland and surrounding tissues.The segmented image was reconstructed by surface drawing for three-dimensional reconstruction of the prostate,and rendered and smoothed by Mimics21 to obtain a high-quality three-dimensional model of the prostate.The augmented reality software and hardware are analyzed,and a three-dimensional visualization system of prostate augmented reality is built to provide a visual scene for prostate puncture surgery.The coordinate conversion relationship of each link of the prostate augmented reality system was analyzed to determine the three-dimensional registration process.The successive ultra-relaxation(SOR)algorithm,Kd-Tree algorithm and singular value decomposition algorithm(SVD)were used to improve the iterative nearest point algorithm(ICP)to study the three-dimensional registration technology,and the registration accuracy of the improved algorithm was verified by simulation experiments,so as to improve the visualization effect of the prostate augmented reality system.The path planning analysis of artificial potential field method in augmented reality environment is carried out,the gravitational potential field and repulsive potential field are improved,and combined with the characteristics of flexible needle puncture,an improved artificial potential field path planning algorithm suitable for flexible needle puncture is proposed.Compared with the artificial potential field method,the artificial potential field method based on simulated annealing and the improved artificial potential field method in this thesis,the simulation path planning experiment is carried out in different obstacle environments,and the improved artificial potential field path planning algorithm is verified to be more in line with the flexible needle puncture in the real environment.The simulation path planning experiment of improved algorithm was carried out on the prostate model,and it was verified that the planned path could effectively avoid obstacles to reach the set tumor location and display it in the augmented reality environment.According to the research content,the experimental materials were selected,the puncture test platform was established,the flexible needle puncture experiment was carried out,and the experimental results were compared with the simulation path planning results to verify that the maximum error size was 1.81 mm and the average error was 1.07 mm,which met the medical puncture error range,and verified the rationality and accuracy of the improved algorithm for flexible needle puncture path planning. |
