节点文献

由核磁共振图像构建头/脑真实模型

Reconstruction Head/Brain Real Model from Mri Image

【作者】 杨雅妹

【导师】 贺士娟;

【作者基本信息】 河北工业大学 , 物理电子学, 2004, 硕士

【摘要】 脑是神经系统中最重要的部分。人类的大脑除了具有和基本的生存相关的功能以外,还具有发现和利用自然界基本规律的能力。而脑科学研究的目的,就是要发现生物的大脑、尤其是人的大脑处理和利用信息的机制。这些机制的发现,将为人类更充分地利用各种信息资源、提高生产效率和生活水平提供前所未有的更为广泛有效的解决办法。 本文主要利用核磁共振成像MRI(Magnetic Resonance Imaging)图像序列,展开了以下几方面的研究: (1)图像的去噪。对图像去噪的方法进行了比较,根据图像的特点,采用了自适应维纳滤波方法,维纳滤波是基于最小二乘滤波原则寻找图像的估值来实现图像滤波的。 (2)体数据的获取。采用灰度阈值技术和数学形态学的操作获取头的数据;在对脑的分割中,本文提出了一种根据脑图像特点和相邻图像在几何结构上具有相似性的特点,构建了自适应模板匹配检测脑的方法:首先选择其中最易处理的图片应用阈值化算法和形态学方法提取出脑轮廓,然后根据相邻图像之间形态具有相似性的特点,再利用形态学算法实现脑体数据分割的操作。在数据的离散上,设计了一种合适的边缘搜索、离散和合理的数据存储结构。 (3)真实头模型的构建。在构建剖分模型上,用基于最短对角线(在两层轮廓中心偏离较大的情况下,对轮廓进行了对中变换)和基于3D Delaunay的方法构建了计算模型;基于表面绘制的思想,实现了头和脑的可视化。 (4)利用扫描线法以及改进的Delaunay方法进行了平面任意区域的FEM三角剖分,二维区域的单元划分是有限元计算的第一步,剖分结果的正确性保证了整个有限元分析结果的可靠性。

【Abstract】 The brain is the most important part of the nerve system. The purpose of the investigation of brain is to discover the principle of the human brain. It’s important for us to use all kinds of information resource.The main aspects of the thesis are as below:(1) Image De-noising: Given some commonly used filter results and the comparisons of them, and choose the wiener filter algorithm. Wiener method is based on statistics estimated from a local neighborhood of each pixel.(2) Acquisition of Head and Brain Volumetric Data and Development Real Head and Brain Reconstruction and Visualization Algorithms: By using threshold technique and morphological operations such as dilate and erosion, head segmented data were obtained. To obtain the brain segmented data, construct an initial template, design and developed a naval algorithm to make the initial template dynamic change with image slice to get the brain data; Designed a framework for searching, sampling and storing contour data.(3) Uniformed contours stacked together and constructed real Boundary Element Method (BEM) calculation models for E/MEG research respectively by minimum distance method and 3D-Delaunary based triangulation growth method. Finally developed visualization algorithm based on surface rendering for the purpose of integration of structure and function information.(4) Using scan line algorithm and improved Delaunay algorithm to generate triangulation meshes. Mesh generation of 2D area is the first step of finite element calculation. The validity of the triangulation guarantees the correctness of the result of the finite element calculation.

  • 【分类号】TP391.41
  • 【被引频次】1
  • 【下载频次】184
节点文献中: 

本文链接的文献网络图示:

本文的引文网络