Level set segmentation method in medical image segmentation research and application
Medical image segmentation can be divided into two categories: one is the region of interest (ROI) identification; the other is the description of the integrity and the extraction of interest region. The emergence of the level set method greatly promoted the development of medical image segmentation. This paper studies three different level set segmentation algorithm to achieve the effective segmentation for brain gray matter and white matter of MRI image. Medical image segmentation is to segment the special meaning areas(such as lesions) into different areas of organizations, so each of the disjoint regions to meet the consistency of specific area. Medical image segmentation is the best application field in the image segmentation. Medical images can be affected, such as noise, field shift effect, the local organizing campaigns and other organizations effect and impact, which resulted in of the complexity and diversity of medical image, thereby greatly increasing the difficulty of image segmentation. Medical image segmentation is highly targeted, various algorithms are often based on specific areas, specific pattern, a particular object, and segmentation results were affected by the impact of subjective factors. Therefore, it is impossible that a segmentation algorithm s to be effective for all types of medical image. The most effective way is combining many different ways to complete segmentation, which set different parameters for different types or component characteristics of the image features and anatomical structure. After years of research, Level Set medical segmentation algorithm combination of ITK and other tools have been further developed, could obtain accurate segmentation results.
LSM image segmentation ITK level set
Miao Yu ShiWeili
Changchun University of Science and Technology, China
国际会议
2011 International Conference on Mechatronics and Applied Mechanics(2011年机电一体化与应用力学国际会议 ICMAM 2011)
香港
英文
1012-1015
2011-12-27（万方平台首次上网日期，不代表论文的发表时间）