喉的数字解剖与虚拟现实仿真

发布时间：2018-03-11 04:19

本文选题：数字解剖学　切入点：数字人　出处：《第三军医大学》2008年硕士论文　论文类型：学位论文

【摘要】： 数字解剖学就是应用数字化的技术手段,对人体的解剖结构及其功能进行数字化,并对数字化的信息进行解剖学观察、测量、虚拟操作等方面研究和应用的一门科学。随着CT、MRI技术的出现与发展,使得人体内部结构的数字化与三维可视化成为可能,特别是美国可视人和中国可视化人体数据集的相继问世,提供了高精度的彩色人体断层图像,为高精度人体数字结构模型的建立提供了理想的数据源。利用数字化人体数据集可以建立人体的数字模型,如人体的几何模型和体素模型,在数字模型的基础上,结合解剖学教学的特点,可以满足断层解剖学、系统解剖学标本的演示等常规解剖学的教学需求,并能实现任意方位的断层图像的获取、无损坏标本示教、远程教学等常规解剖学研究和教学中无法实现的目的。目前数字解剖学还处在初始阶段,在数字化人体数据的获取过程中,在图像分割、三维模型的建立与虚拟仿真中,以及异地远程教学方面仍存在许多问题,需要进一步完善与提高。为此本实验拟在现有的数字化人体建模技术的基础上,从数据获取与处理、图像分割与三维重建、模型可视化与虚拟仿真等方面探讨数字解剖学教学实现的技术方法,为数字解剖学提供理论依据和技术平台。本实验选择第二例中国数字化可视人体数据集的喉区断层图像(下颌角与甲状腺下缘之间)作为主要研究对象,同时通过冰冻铣切技术获得冠状、矢状位各1例头颈部数据集,根据图像上配准点的特征进行自动配准,建立相应的高精度配准数据集。在图形工作站上利用Amira4.1软件对3例数据集进行三维重建,利用数字化重采样技术对建立的体素数据体进行数字化虚拟切割,从多方位模拟同一例标本断层图像的获取。采用手工的分割方法对第二例中国数字化可视人体数据集喉区的结构进行精细分割,利用自行开发的软件实现分割数据格式的转换,使得一次图像分割的结果可用于多种可视化软件系统。利用体绘制和面绘制的方法在Dextrobean和Amira虚拟现实系统中实现了喉区重建结构的三维立体显示,模拟喉的解剖学教学和临床技能的仿真操作,探讨高精度图像分割的方法,以及与数字解剖学的相关的虚拟现实仿真技术与软硬件平台建设。将Amira软件系统中建立的面模型经模型简化与平滑处理后输出为VRML模型,利用VRML模型的可视性、交互性,结合HTML和JavaScript语言对喉区的数字模型进行基于网络的虚拟解剖学教学研究,建立了喉的数字解剖学网络教学系统,并探讨数字解剖学三维立体模型在网络的传输与虚拟仿真操作。其研究成果有助于数字解剖学的研究和开展,为数字解剖学的教学提供技术支撑和应用平台。本研究的主要研究结果和结论如下: 1.根据冰冻断层图像上的配准点特征信息,利用自主开发的软件实现了数据集图像的自动配准,其配准误差小于1个像素,达到亚像素水平。 2.利用数字化重采样技术对横、冠、矢3例颈部的数据集分别进行了同一例标本多方位断层图像的采集与观测,实现虚拟断层解剖学的研究与观察,为临床影像诊断和数字化虚拟断层解剖学的开展提供了技术方法和形态学资料。 3.利用手工分割的方法获得了喉区结构的高精度分割数据集,从图像分割到三维数字化建模与虚拟现实仿真等方面进行技术上的探讨,建立了一套完整的技术路线。用自制的分割数据集格式转换软件实现了分割数据精确转换,使得一次分割的结果可用于多种可视化软件系统。 4.采用面绘制和体绘制的算法建立了2种喉的精细数字化解剖模型,可作为解剖学教学标本的补充。并对建立的喉模型进行虚拟的解剖学操作,为数字解剖学的进一步发展奠定了基础。 5.建立了喉区结构的VRML数字模型,通过事件路由设置实现了模型的可视化控制,结合HTML和JavaScript语言实现基于网络的喉区数字解剖模型的远程读取,本地浏览和交互式操作,为解剖学的远程虚拟仿真教学提供技术平台和方法,有助于远程虚拟数字解剖学的研究和开展。
[Abstract]:Digital anatomy is the application of digital technologies, digital to human anatomical structure and function, and anatomical observation, the digital information measurement, a scientific research and application of virtual operation etc.. With CT, the appearance and development of MRI technology, which makes it possible to digitize and three-dimensional visualization of the internal structure of the human body. Especially the visible human and Chinese visible human data sets have been developed, provides high precision color human images, the establishment of high precision digital human body model provides an ideal data source. Set can build digital models of the human body by using digital human data, such as geometric model and human body model, in the basis of the digital model, combined with the characteristics of anatomy teaching, can satisfy the sectional anatomy, anatomic demonstration system of conventional anatomy The teaching needs, and can achieve any range of image acquisition, no damage samples of teaching, to realize the remote teaching anatomy in the research and teaching purposes. The digital anatomy is still in the initial stage, in the process of acquiring digital human data, image segmentation, and the establishment of virtual simulation of 3D model. There are still many problems and remote teaching, the need to further improve and improve.
This experiment is based on digital human modeling technology on the current from the data acquisition and processing, image segmentation and 3D reconstruction, model visualization and virtual simulation aspects of technology to realize digital anatomy teaching, provide a theoretical basis and technical platform for digital anatomy. In this experiment, second cases of laryngeal region image digital Chinese the visible human data set (with the lower edge of the mandibular angle between the thyroid) as the main research object, and through the milling technique for coronary and sagittal 1 cases of head and neck data set, according to the image feature registration point for automatic registration, the establishment of high precision registration data corresponding to the set of Amira4.1 software for three-dimensional. The reconstruction of 3 cases of data sets used in graphics workstation, using digital resampling technology in data to establish the digital virtual cutting, from multiple angles The simulation for the same samples. Using image segmentation method for fine segmentation of hand structure of second cases of laryngeal region set China visible human data, convert data format segmentation using self-developed software, the results of image segmentation at a time can be used for a variety of visualization software system. By using the method of volume rendering and surface rendering three-dimensional reconstruction of laryngeal region structure in Dextrobean and Amira in virtual reality system, simulation operation simulation of laryngeal anatomy teaching and clinical skills, to explore the method of high accuracy of image segmentation, and virtual reality simulation technology and digital anatomy and platform construction of software and hardware. The surface model of Amira software in the system by using simplified model and smoothing processed into VRML model, using the visual VRML model, interactive, combined with HTML and JavaScript. Digital model of laryngeal region of virtual anatomy teaching based on network, established a network teaching system of digital anatomy of larynx and discuss digital anatomy, three-dimensional model in network transmission and virtual simulation. The research results have contributed to research and development of digital anatomy, provide technical support and application platform for digital anatomy the teaching.
The main results and conclusions of this study are as follows:
1., according to the feature information of registration points on frozen fault images, automatic registration of dataset images is realized by self-developed software. The registration error is less than 1 pixels, reaching sub-pixel level.
2. the use of digital resampling technique on the cross, crown, 3 cases of cervical sagittal data sets were collected and observed the same specimen range images. Research and implementation of virtual anatomy observation, provides the technical method and morphological data for clinical imaging diagnosis and virtual anatomy.
Methods 3. using manual segmentation obtained throat zone structure of high precision segmentation data set from image segmentation to 3D digital modeling and virtual reality simulation and other aspects of technology, the establishment of a complete set of technical route. The format conversion software to realize the segmentation and accurate data conversion with self-made segmentation data set, the the results of segmentation time can be used for a variety of visualization software system.
4., by using the algorithm of surface rendering and volume rendering, 2 kinds of fine digital anatomical models of larynx have been established, which can be used as a supplement to anatomical teaching specimens. The virtual anatomy operation of laryngeal models has been established, which laid the foundation for further development of digital anatomy.
5. the establishment of a VRML digital model of laryngeal region structure, through the event routing set to achieve the control of model visualization, combined with HTML and JavaScript language to read remote network digital anatomical model of laryngeal region based on local browsing and interactive operation, provide technical platform and simulation method for the remote virtual anatomy teaching, research and development help the remote virtual digital anatomy.

【学位授予单位】：第三军医大学
【学位级别】：硕士
【学位授予年份】：2008
【分类号】：R322;R319

【相似文献】