Image-guided Surgery In Treatment Of Necrosis Of Femoral Head By Core Decompression

The technology of computer-aided surgery involves the field of mathematics, computer graphics, mechanical engineering and biomedical engineering, which has become a hot research area of modern surgery. It is used for the development of pre-operative surgical planning and intraoperative navigation, tracking surgical instruments in the course of operation, and real-time display of the relative position between the patients and instruments, hence to provide doctors with rich navigational information. It can also record and store the path of the instruments, measure the implants angle, length, diameter and implement accurate operation, rather than relying on experience to solve the problem. At the same time, it reduces the dependence of orthopedic surgeons on the use of C-arm X-ray machine in the traditional operation, and the sufferings caused by X-ray radiation.Being more and more common, osteonecrosis of the femoral head has drawn a wide range of medical attention. One of the treatment called core decompression, which is simple, small invasive and easily recovered, has good effects in the early stage. But doctors have to use X-ray machine on patients with continuous exposure to obtain two-dimensional X optical imaging to determine the spatial location of patients and instruments during the operation. Without three-dimensional image guiding it is difficult to judge the surgical instruments in space; it not only waste a great deal of time in the process of surgery but also increase the impact of doctors and patients because of X-ray irradiation. In this paper, we take consideration of the limitations of traditional treatment, researched image-guided surgical technology and related mathematical methods. Based on these we developed the surgical navigation system mainly for the treatment of osteonecrosis of the femoral head. The detailed work include: 1. Discussed in detail with the image processing and three-dimensional visualization technology related to volume data. In pre-imaging, introduced some methods of image filtering and interpolation and used linear interpolation to improve the accuracy of image reconstruction. In image segmentation, realized the image segmentation based on gray and region. In the 3D reconstruction, we discussed the Marching Cubes iso-surface extraction algorithm, as well as the solutions to the problem of ambiguity and realized the 3D reconstruction of the femur. Finally, we utilized Garland's method for mesh simplification, reduced a lot of computing and made it possible for model cutting or clipping.2. Given the problem of registration between patient and model, a method was brought forwarded, which is a two-step way. The first is an initial estimation using least squares method and the next is an accurate one named iterative closest point algorithm. In the pre-operative planning stage, we adopt the Z Buffer technology to obtain the point set of the external surface of the 3D femur model as the input data for registration, in place of the model contained points within it, which increased registration accuracy.3. The commonly used method in CAD/CAM, virtual collision detection technology was adopted and applied to the surgical navigation, to determine the collision between models, and then make it possible to cut or clip models in real time. We also discussed the implicit function approach. By combining simple geometric shapes to build a model which is similar to the operation tool, we achieved the model cutting. By this the doctor can get more intuitive understanding of the progress of the surgery.