Research on virtual three dimensional reconstruction system for medical slices image

Traditional medical reconstruction system mainly aims at processing medical CT images with clear boundaries. As for those sequence slice images with subtle and unclear boundaries, manual segmentation algorithm are usually adopted. However, they reduce the efficiency greatly. In this paper, we have conducted a series of deep study on these issues, and propose an image segmentation method which combines the algorithm of region growing and linear interpolation method between images. At the same time, research on the accelerating algorithm of three-dimensional (3D) reconstruction has also been conducted, and a 3D spatial filtering method is proposed to filter the 3D reconstruction model. The experimental results show that with the methods mentioned above, the reconstruction effect is improved, and its applicability becomes broader and efficiency becomes higher.