CASIA OpenIR  > 毕业生  > 硕士学位论文
Alternative TitleThe Design and Implementation of Segmentation Platform for Medical Image
Thesis Advisor田捷
Degree Grantor中国科学院研究生院
Place of Conferral中国科学院自动化研究所
Degree Discipline模式识别与智能系统
Keyword医学图像 图像分割 算法开发包 插件 Medical Image Image Segmentation Toolkit Plugin
Abstract医疗诊断设备极大地提高了人类获取自身信息的能力,推动了现代医学的 发展。这些设备获得的信息主要以图像的形式存在,因此对这些图像的处理就 成了充分发挥其效果的关键。由于其成像方式的共同特点,我们把这些来自各 种医疗诊断设备的图像统称为医学影像。 随着科技的发展,在医学影像技术研究的各个领域都出现了很多新方法, 本文主要针对医学影像处理中的分割算法,设计与实现了医学影像分割算法平 台,其目的在于:为分割算法的研究人员提供一个算法集成的平台以避免重复 工作;为工程技术人员提供一个接口统一、灵活易用的算法包:为医务人员提 供一个选择多样、简单易用的应用程序。 本文首先概要介绍了医学图像、以及医学图像分割的特点,然后介绍了医 学影像分割平台的总体框架设计及设计准则,最后重点描述了各个分割算法包 及其插件的设计与实现。本文主要有如下贡献: 11应用设计模式的思想,提出了一种开放的、统一的分割算法开发包设计 框架; 2)实现了部分主流分割算法。目前分割算法包中提供的分割算法主要有: fast marching算法、level set算法、live wire算法、种子点生长算法、交 互式分割算法和闽值分割算法; 31实现了部分分割插件。分割算法包为用户提供了一个开发包,而与其相 对应的分割插件贝lJ为分割过程提供了一个用户界面和人机交互平台。目 前分割算法平台主要提供了与分割算法开发包中的分割算法相对应的 分割插件; 目前,医学影像分割算法开发包已经被集成在我们实验室近期研发的集成 化医学影像算法平台MITK(Medical Image Toolkif)中;与分割算法包相对应的分 割插件也集成在了三维医学影像处理平台3DMed中,它们都已经在:Intemet网 上面向国内外免费发行,相关科研人员可以下载并进行二次开发。
Other AbstractThe medical image equipment has greatly increase human being's ability to gain information of themselves, and it also push the improvement of modern medicine. The information these equipments achieved is mainly stored as images, thus how to process these images becomes a key problem. Because of their common ways of imaging, we call all the images coming from a variety of medical equipment as medical images. With the development of sience and technology, there emerge many new methords in every field of the medical image research. This paper mainly aims at the segmentation methord of medical image processing. The author design and implement the segmentation platform of medical images. The gola of the platform is to: provide an integrated platform for researchers to avoid redundant work; provide a flexible toolkit of united interface for the engineers; provide an application program for the doctors so that they can choose the segmentation methord they want. At the firstly, the paper give a brief introduction to the charictistics of medical image and the difference between the segmentation of medical images and the general images. Then the paper describes how the famework of the platform is designed and the design principles. At the end, the paper emphasize on the implementation of each toolkit and their plugin. The paper's contribution is: 1) Put forward an open and unitied framework for segmentation toolkit by applying several design patterns; 2) Implement some segmentation arithmetics. They are, fast marching methord, level set methord, live wire methord, region grow methord, interactive methord and threshold methord. 3) Implement some segmentation plugins. While the segmentation toolkit provides a toolkit for the clients, the segmentation plugin provides a window and a user interface for the clients. The paper implements the segmentation plugins according to the segmentation methords provided by the segmentation toolkit. The segmentation toolkit of medical images has been integrated in the Medical Image Toolkit (MITK) which is developed by my lab; The plugins are also integrated in the 3DMed platform. Both of them are now published on the internet. Related scientitsts can download for free.
Other Identifier759
Document Type学位论文
Recommended Citation
GB/T 7714
朱珣. 医学影像分割算法平台的设计与实现[D]. 中国科学院自动化研究所. 中国科学院研究生院,2004.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[朱珣]'s Articles
Baidu academic
Similar articles in Baidu academic
[朱珣]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[朱珣]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.