医学图像分割简介及其应用实例.pptVIP

医学图像分割简介及其应用实例 自我介绍 杨妍, 水木ID – recky; 水木老虫子 ? - 99年初注册ID; 本科：清华生医8；跟高小榕教授做毕设；2002年本科毕业； 目前：美国 Georgia Institute of Technology (Georgia Tech / GT) 博士三年级；Biomedical Engineering Department, Bioengineering Program; Cardiovascular lab (Dr. Don P. Giddens); Minerva lab (Dr. Allen R. Tannenbaum). 实验室简介 Cardiovascular lab (Dr. Giddens) 利用计算流体力学结合医学成像的方法进行血流分析：血流速度、压力、剪切力、剪切力剃度等重要参数与血管动脉硬化的分布之间的关系。 医学成像：获得血管的几何形状（CT）和血液进口出口的速度（Phase Contrast - MRI）等； 计算流体力学：利用图像里所获得的信息进行血流分析，得到精度较高的参数分布，以便分析血流与病灶之间的关系。 实验室简介（2） Minerva lab (Dr. Allen R. Tannenbaum) 各种图像处理技术– 静态和动态图像；医学和军事图像； 医学：图像分割，虚拟内窥镜，三维人体器官的展平和可视化，超声图像的去躁等； 军事：基于活动轮廓和level-set的动态目标（如飞行器）有哪些信誉好的足球投注网站，Kalman filter和particle filter的应用，控制理论等。 Introduction to Image Segmentation - Definition Classically, image segmentation is defined as the partitioning of an image into non-overlapping, connected regions that are homogeneous with respect to some characteristic such as intensity or texture. Pixel segmentation, rather than classical segmentation is often a desirable goal in medical images, particularly when disconnected regions belonging to the same tissue class. When the constraint that regions be connected is removed from the classical segmentation, determining the classes is called pixel classification. Image Segmentation Methods1. Thresholding Thresholding approaches segment scalar images by creating a binary partitioning of the image intensities; Simple yet often effective; Usually used as an initial step in a sequence of image processing operations; Does not take into account the spatial characteristics of an image; Sensitive to noise and intensity inhomogeneities. Image Segmentation Methods2. Classifiers and Clustering Methods Classifiers and clustering methods are pattern recognition techniques that apply to image segmentation problems; Classifiers are known as supervised methods because they require training data; Clustering algorithms are termed unsupervised because they perform classification of the data without the u