节点文献
基于遗传算法的肝CT序列图像分割应用研究
【作者】 刘兵全;
【导师】 何继善;
【作者基本信息】 中南大学 , 生物医学工程, 2004, 硕士
【摘要】 计算机层析成像(CT)被广泛的应用于肝脏疾病的诊断,成为医生进行肝功能分析,病理和解剖研究的重要手段之一,设计和发展图像处理技术以提高医生的诊断效率和水平成为了近年来一个重要的研究领域。 图像分割是指将图像中具有特殊含义的不同区域区分开来的一种技术。分割后的子区域是互相不交叉的,每一个区域都满足特定区域的一致性。图像分割在医学应用中具有特殊的重要意义,它是医学图像处理的关键一步,是一个跨越医学和计算机科学的综合性研究课题。 肝脏边界的确定是定量分析的前提,原先医生必须对肝CT序列中的每一幅图像都要手工确定肝脏边界,这耗时费力,效率也不高。因此,研究肝脏边界的自动提取技术成为迫切的问题。图像的边界提取方法很多,但效果都不佳,本文在分析了肝CT序列特点基础上,利用遗传算法搜索边界,取得了较好结果。 首先,介绍了图像分割的定义、意义以及国内外医学图像分割的发展现状和趋势,并在此基础上提出了基于遗传算法的肝CT序列图像分割的设想。 其次,概述了当前主要的医学图像分割算法;遗传算法的基本思想,实现过程和特点。 再次,本文根据肝CT序列图的特点,采用遗传算法对肝CT序列图进行肝脏边界轮廓提取,设计了适当的遗传算法编码方式和适应度函数以及遗传操作算子。 在论文最后,对整个论文的工作进行了总结,并指出了进一步研究探索的方向。
【Abstract】 Computed tomography(CT) images have been widely used for liver disease diagnosis. Designing and developing image processing techniques to help doctors improve their diagnosis has been a research hot over the past years.Image Segmentation is a kind of image processing technique that devide an image into several subareas. Those areas are not intersectant and each has its own property. Image Segmentation is a key step in image processing. It is also a synthetic research task which is relate to medical and computer science. With the rapid development of medicine, Image segmentation is taking an important role in medical application.The detection of the liver contour is the foundation of quantitative analysis. Originally, doctor had to do it by himself from first image to the last. It was inefficient and needed a lot of time. So, how to use computer technology to extract liver contour automatically is an important research field. Many methods are used in edge extraction, but they both had disadvantages. After analysis the characteristics of CT liver serial images, using genetic algorithms(GA) to extract liver contour. It is proved to be feasible and effective.Firstly, The definition of Image Segmentation, its meaning and Trend of development are enumerated. And based on this, genetic algorithms is bring forward.Secondary, Summarized several popular medical image segmentation methods. Summarized the base theory of GA, implementing processes and its characteristics.Thirdly, According to the characteristics of serial CT liver images, GA is used in extracting the contour of Liver. Selected suitable coding mode, designed fitness function and genetic operators- reproduction, crossover and mutation.In the last part of this article, the author’s ideas and advices to extend study are given.
【Key words】 CT; liver; image segmentation; genetic algorithms; edge extraction;
- 【网络出版投稿人】 中南大学 【网络出版年期】2004年 04期
- 【分类号】R-39
- 【被引频次】12
- 【下载频次】407