Research And Implementation Of Liver Segmentation And Visualization In CT Images Based On Anatomical Knowledge

The general segmentation problem is the process of partitioning an image or data-set into a number of uniformity or homogeneous segments. Medical image segmentation is a fundamental problem in medical image processing and analysis, and is the basis of computer aided diagnosis and treatment as well, for example,3D Visualization, Computer Aided Operation, and Radiology Treatment all assume that Region of Interests (ROI) are well segmented.Hepatic disease such as viral hepatitis, fatty liver, hepatocirrhosis and hepatocarcinoma is one of the main diseases that threat us in current society. Liver operation like liver transplant has become popular in treating hepatic disease in many countries. CAD (computer-aided diagnosis) and pre-surgical planning of hepatic disease become an urgent need. Accurate segmentation of liver as the foundation of them is becoming a research focus in the medical imaging segmentation area which has a wide range of application prospects and practical significance.This thesis studies the semi-automatic and interactive segmentation of liver using multi-slice CT Images. The difficulties of liver segmentation are described as follows: Firstly, the surrounding tissues have so similar intensity with the liver that it is difficult to separate them based on the difference of intensity. Secondly, partial volume effect blurs the outline of the liver decreasing the reliability of liver segmentation. Finally, the shape of liver differs from person to person and it may be different for the same person.Level set method is a popular and widely used algorithm in present image segmentation area. Its basic idea is to construct an energy function for the model and let the curve evolve under the model’s inner control force and outside image force. When the energy function reaches its minimum value, the evolving curve will describe the target region. However, there are still some shortcomings in level set methods, such as under-segmentation, over-segmentation and leakage problems.To avoid the shortcomings of existing liver segmentation algorithms, author proposes a semi-automatic method based on a special level set method namely DRLSE model. This segmentation mainly includes three stages:pretreatment is applied to the original images at first including intensity adjusting and median filtering; then the coarse segmentation result is got by applying region growing method; the fine liver contour can be got by DRLSE model. Experimental results demonstrate that DRLSE method is15times faster than traditional CV method. As for the accuracy, the former is also much better than the latter. Besides, the paper uses Marching Cubes algorithm to visualize the segmentation results based on VTK.