Computer-aided Orthopedic Surgery System Based On Bone Tissue Segmentation And Augmented Reality Technology

With the rapid development of medical imaging and computer technology,the computer-aided orthopedic surgery systems based on medical image have played an important role in clinical practice.At present,the pre-orthopedic examination and intraoperative positioning can only be performed through the computer-aided orthopedic surgery system for reading and analysis so that the doctors form a subjective impression in their brain,and they locate the surgical lesion area of the patient relying on rich clinical experience.The accuracy of the surgical positioning is very dependent on the doctors’ clinical experience and the information can not be intuitively transmitted to other members of the team.The application of augmented reality in the computer-assisted orthopedic surgery system can assist the doctors to segment the CT image of the patient’s surgical site and locate the patient’s lesion area in orthopedic surgery,so that doctors can share the surgical information with other members in real time and improve the accuracy and speed of surgical positioning.In order to efficiently assist doctors in the pre-orthopedic examination and intraoperative positioning,this paper designs a computer-assisted orthopedic surgery and augmented reality system based on image segmentation technology,three-dimensional visualization technology and augmented reality technology.The research content of this paper is mainly divided into the following four aspects:1.In order to accurately extract the relevant bone tissue for the research of computer-aided orthopedic surgery system,this paper studies the bone tissue segmentation algorithm based on medical image segmentation technology.First,the reading,decoding and preprocessing of DICOM images are introduced.Then,a coarse segmentation method of bone tissue based on adaptive dual-threshold is adopted to segment out the pixels of suspected bone tissue.Finally,the three-dimensional region growth method is used to divide the suspected bone tissue to obtain the complete bone tissue.2.In order to further implement the segmentation between bones,an automatic bone segmentation method based on Graph-Cut algorithm is adopted.First,the bone image is sharpened to enhance the details and edge information.Then,an improved 3D structure enhancement filter based on Hessian matrix is adopted,and the filter result is added to the input feature of the Graph-Cut algorithm.Then the morphological erosion operation is used to separate the adjacent bones that are stuck together.Finally,the Graph-Cut algorithm based on the maximum flow and the minimum cut is used to segment the bones,so that the automatic bone segmentation is implemented.3.In order to implement the Augmented Reality technology in the computer-assisted orthopedic surgery system,this paper proposes a method to implement augmented reality based on image feature point matching and 3D visualization.First,the results of bone segmentation are reconstructed by using the Marching Cubes surface rendering algorithm to obtain a virtual 3D model for implementing augmented reality technology.Then,a video image matching algorithm based on the improved SURF feature points is used to track the target in the video.Afterwards,the Perspective-N-point problem is used to solve the spatial pose of the video target,and the virtual three-dimensional model is rendered in the video image with the corresponding pose to achieve the Augmented Reality effect of the bone model.4.Based on the research contentabove,this paper independently designed and developed a computer-aided orthopedic surgery system based onbone tissue segmentation and augmented reality technology.The system is mainly divided into two modules: medical image processing module and augmented reality module.The function of the medical image processing module is mainly to segment bone tissue from the bone CT image and three-dimensional visualization of the segmentation results,including interactive bone segmentation,automatic bone segmentation,3D visualization of the segmentation results and so on.The function of the augmented reality moduleis to implement the augmented reality technology,including video processing,video target tracking,three-dimensional model space pose control and other functions.The system can be applied to the precise positioning of the lesion area in orthopedic surgery to assist the doctors’ examination and positioning.The experimental results show that the bone tissue segmentation algorithm and the inter-bone segmentation algorithm adopted in this paper have good results in visual assessment and quantitative assessment,and the computer-aided orthopedic surgery system developed in this paper can be applied to the positioning of clinical orthopedic surgery to assist doctor carry out scientific inspection and positioning.