CT Image Segmentation And 3D Reconstruction

Nowadays,image processing technology can be seen everywhere in our lives and plays an irreplaceable role.In medicine,every moment in the world,doctors are using medical images to help patients diagnose the disease.The emergence of medical image technology greatly reduces the difficulty of diagnosing the disease and improves the medical level.According to the statistics of the World Health Organization,388000 people die of liver cancer every year.Early detection and treatment are the best way to reduce mortality.The existing treatment methods are inseparable from the accurate segmentation of the liver and liver.However,the fuzzy edge,shape and size of the liver and liver tumor will vary from person to person,which makes the segmentation difficult.In clinic,the segmentation of liver and liver tumor mostly depends on the manual description of medical staff.Manual segmentation is tedious and inefficient.Finding more intelligent segmentation algorithm and replacing artificial with machine has become a research hotspot of many scholars.The segmentation algorithm in this paper is mainly based on the level level algorithm,and the level set algorithm is a kind of active contour model,which has the following advantages:(1)When the initial contour is set,the prior information of human,such as size and shape,can be integrated to improve the efficiency of the algorithm.(2)The level set method can transform the complex image segmentation problem into the curve evolution problem,because the curve itself is closed,which ensures the continuity of segmentation results and converges to the target edge.In terms of liver segmentation,this paper takes the level of distance regularization as the framework,and achieves good segmentation results by adding region information to the energy function.In the segmentation of liver tumor,this paper takes the regional statistical active contour model as the framework,and adopts different scale parameters in different regions,which effectively overcomes the problem of uneven gray distribution.Bycalculating the segmentation performance evaluation index of the above two segmentation algorithms,The reliability of this algorithm is proved by specific numerical value.In the initial contour sensitivity verification,different initial contours get similar segmentation results,which proves that the algorithm in this paper is not sensitive to the initial contour.Finally,according to the segmentation results of the algorithm,this paper successfully carries out three-dimensional reconstruction of liver tumor based on MATLAB.