基于口腔CT重建数据的牙齿分割与可视化研究

发布时间：2018-12-14 11:29

【摘要】：近些年来,各种医学成像技术飞速发展,衍生出了各种以此类技术为基础的计算机辅助诊断系统,针对不同目的的计算机辅助系统的研究成为该领域的热点课题。低剂量口腔CT正是针对牙科以及口腔颌面部的保健检查和疾病诊疗而设计的计算机诊断辅助系统,而在口腔CT的工作过程中,图像分割和三维可视化是实现检查以及诊疗疾病的关键技术。图像分割一直是一个传统而具有挑战性的课题,是计算机辅助分析、医学影像可视化以及疾病诊断的前提和基础。对于医学图像分割问题,一般可分为两个部分：一是图像中特定目标的识别；二是目标区域完整性的描述和提取。相对于一般图像,医学图像的多样性和复杂性,使得传统的分割方法很难得到理想的分割结果,而借鉴一些流体中的重要思想,Osher和Sethian首次提出了水平集算法,这是一种有效解决曲线演化问题的数值方法,并且计算稳定,适合任意维度空间。水平集方法的出现,极大地推动了主动轮廓模型的发展,水平集方法和曲线演化模型相结合,克服了传统Snake模型的诸多固有缺陷,大大拓展了主动轮廓模型的应用领域。针对牙齿结构的多目标特性,本文主要研究了基于Chan-Vese模型的水平集方法,通过对传统的水平集方法和无需重新初始化的水平集方法的研究,将其应用到口腔CT重建图像中,逐层的提取出目标区域(牙齿),为后续的图像可视化做准备。医学图像的可视化技术是计算机图形学和图像处理在生物医学工程中的重要应用,医学上能将人体的组织和病变部位以三维形式显示出来,为医生的临床诊断和治疗提供帮助。医生可以无损的获得人体内部解剖结构的大量信息,获取各组织器官的相对空间关系,极大地促进了医学的发展。本文主要研究了Marching Cubes算法(MC)的基本原理,并应用其重建基于口腔CT的牙齿三维结构。Marching Cubes算法提出了一种提取等值面和检测体元的方法,是使用比较广泛的三维重建算法。我们针对牙齿的相应结构,对Marching Cubes算法进行适当的调整,能够得到较好效果的牙体三维结构。
[Abstract]:In recent years, various medical imaging techniques have been developed rapidly, and various computer-aided diagnosis systems based on these techniques have been derived. The research of computer-aided systems for different purposes has become a hot topic in this field. Low dose oral CT is designed for dental and oral and maxillofacial health care and disease diagnosis and treatment of computer diagnostic assistance system, and in the oral CT process, Image segmentation and three-dimensional visualization are the key technologies to detect and treat diseases. Image segmentation is a traditional and challenging subject, which is the premise and foundation of computer aided analysis, medical image visualization and disease diagnosis. The medical image segmentation problem can be divided into two parts: one is the recognition of specific targets in the image and the other is the description and extraction of the integrity of the target region. Compared with general images, the diversity and complexity of medical images make it difficult for traditional segmentation methods to obtain ideal segmentation results. For the first time, Osher and Sethian have proposed level-set algorithms for reference to some important ideas in fluids. It is an effective numerical method to solve the problem of curve evolution and is stable and suitable for arbitrary dimensional space. The emergence of the level set method has greatly promoted the development of the active contour model. The combination of the level set method and the curve evolution model overcomes many inherent defects of the traditional Snake model and greatly expands the application field of the active contour model. Aiming at the multi-objective characteristic of tooth structure, the level set method based on Chan-Vese model is studied in this paper. By studying the traditional level set method and the level set method without reinitialization, the level set method is applied to the reconstruction of oral CT image. The target area (tooth) is extracted layer by layer to prepare for the subsequent image visualization. The visualization technology of medical image is an important application of computer graphics and image processing in biomedical engineering. In medicine, the tissues and pathological parts of human body can be displayed in three-dimensional form, which can be helpful for the clinical diagnosis and treatment of doctors. Doctors can obtain a great deal of information about the internal anatomy of human body, and obtain the relative spatial relationship between tissues and organs, which greatly promotes the development of medicine. In this paper, the basic principle of Marching Cubes algorithm (MC) is studied, and a method of extracting isosurface and detecting volume elements is proposed by using. Marching Cubes algorithm of dental 3D structure reconstruction based on oral CT, which is a widely used 3D reconstruction algorithm. According to the corresponding structure of teeth, the Marching Cubes algorithm can be adjusted properly to get a better effect of the three-dimensional structure of the tooth.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：R816.98;TP391.41

【共引文献】