图谱法脑部MRI图像自动分割技术发展及应用..docx

图谱法脑部MRI图像自动分割技术发展及应用何小海1,梁子飞1,唐晓颖2,3,滕奇志1四川大学电子信息学院，成都，610065；2.中山大学卡内基梅隆大学联合工程学院，广州，510006；3.广东顺德中山大学卡内基梅隆大学国际联合研究院，顺德，528300）摘要：脑部MRI图像自动分割是计算机技术运用于医学上的一个典型工作，脑部图像分割技术对于人类研究脑部疾病具有重要意义。虽然有一些非常广泛应用的图像分割方法如阈值法、区域增长法、聚类法等，但在脑部MRI图像分割中，这些方法都没有图谱法具有更为实际的医学研究与临床价值。本文回顾了脑部图像配准、分割的发展历程，介绍了图谱分割算法的发展及基本原理，以及当前比较前沿的多图谱分割系统的组成和应用。最后本文对图谱法脑部自动分割在实际临床医学中的应用前景作了总结和展望。关键词：MRI；图谱法；脑图像自动分割；微分同胚；脑部图像配准中图分类号: TP391文献标志码: AAtlas-based Brain MRI ImageSegmentation Technology Development and Application，Xiaohai He1，Zifei Liang1，Xiaoying Tang2,3，QizhiTeng(1. College of electronics and information engineering, Sichuan University, Chengdu, 610065; 2.SYSU-CMU Shunde International Joint Research Institute, Guangzhou, 510006; 3.SYSU-CMU Shunde International Joint Research Institute, Deshun, 528300)Abstract:Automated segmentation of brain MRI images is an important computer-based technology with wide applicabilityto medicine field, and this technology is of great significance in the study of human brain diseases. There exists a variety of segmentation methods, such as the thresholding method, the region growing method, and the clustering method,that have been broadly applied to natural images. However, those methods are not as powerful or practical as atlas-based method when applied to clinical medical image. This article is to review the development of the atlas-based method forbrain image segmentation, with representative algorithms introduced. The basic principles of these parcellation algorithms aredescribed as well asthe componentsof astate-of-the-art segmentation system. On this basis, this article introduces the segmentation procedures, and discusses its various applications in clinical research. At the end of this paper, we summarize the current status and the future potential of the automated segmentation’s applications in real clinical research.Key words: MRI; Atlas-based; automated brain segmentation; diffeomorphic; brain image registration引言在人类对脑部结构认识过程中已经将脑部划分成很多个功能区，各个区域对人的语言、运动、记