手术导航系统中空间配准算法的研究
发布时间：2019-06-08 11:26
【摘要】：现在随着物质生活水平的不断提高,人们对自身的健康状况也更加的关心,同时对医疗手段也提出了更高的要求。手术导航系统结合了计算机、图像处理和空间定位等多种先进的科学技术,使一些使用传统手术方法无法完成或很难完成的手术成为可能。手术导航系统具有精度高和微创化的特点,相比以往,医生可以在手术前使用手术导航系统进行模拟手术,这些都大大的提高了外科手术的安全性。手术导航系统符合了人们对新的先进医疗技术的要求。本文的主要研究内容是手术导航系统的关键技术之一空间配准技术,空间配准就是将患者病灶所在的空间坐标系和虚拟空间坐标系统一在一个坐标系内。本文分别对成对点集的空间配准和非成对点集的空间配准进行了研究。对于成对点集本文首先使用传统的四元数算法和SVD矩阵分解法进行配准,而对于非成对点集采用最近点迭代算法(Iterative Closest Point,ICP)算法进行配准。为了提高空间配准的精度,针对成对点集的空间配准,本文结合了ICP算法对四元数算法进行了改进;针对非成对点集的空间配准,本文结合了Delaunay剖分对ICP算法进行了改进。最后本文对提出的两种改进算法进行了仿真实验。通过对实验结果的分析可知,两种改进的配准算法在运算上花费的时间比之前要长,但是仍在可接受的范围之内,与此同时在配准精度上有了明显的提高,达到了预期的效果。
[Abstract]:Now, with the continuous improvement of material living standards, people are more concerned about their own health, and put forward higher requirements for medical means. Surgical navigation system combines computer, image processing and spatial positioning and other advanced science and technology, making it possible to use traditional surgical methods can not complete or very difficult to complete the operation. Surgical navigation system has the characteristics of high precision and minimally invasive. Compared with the past, doctors can use surgical navigation system to simulate surgery before operation, which greatly improves the safety of surgery. The surgical navigation system meets the requirements of new advanced medical technology. The main research content of this paper is one of the key technologies of surgical navigation system. Spatial registration is to put the spatial coordinate system and virtual spatial coordinate system in one coordinate system. In this paper, the spatial registration of pairwise point sets and the spatial registration of non-pairwise point sets are studied respectively. In this paper, the traditional quaternion algorithm and SVD matrix decomposition method are used to register the pairwise point set, while the nearest point iterative algorithm (Iterative Closest Point,ICP) algorithm is used for the non-pairwise point set. In order to improve the accuracy of spatial registration, this paper improves the quaternion algorithm combined with ICP algorithm for the spatial registration of pairwise point sets, and improves the ICP algorithm with Delaunay partition for the spatial registration of non-pairwise point sets. Finally, two improved algorithms are simulated. Through the analysis of the experimental results, it can be seen that the operation time of the two improved registration algorithms is longer than before, but it is still within the acceptable range. At the same time, the registration accuracy has been obviously improved, and the expected results have been achieved.
【学位授予单位】：河北工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：R616;TP391.41
本文编号：2495248
[Abstract]:Now, with the continuous improvement of material living standards, people are more concerned about their own health, and put forward higher requirements for medical means. Surgical navigation system combines computer, image processing and spatial positioning and other advanced science and technology, making it possible to use traditional surgical methods can not complete or very difficult to complete the operation. Surgical navigation system has the characteristics of high precision and minimally invasive. Compared with the past, doctors can use surgical navigation system to simulate surgery before operation, which greatly improves the safety of surgery. The surgical navigation system meets the requirements of new advanced medical technology. The main research content of this paper is one of the key technologies of surgical navigation system. Spatial registration is to put the spatial coordinate system and virtual spatial coordinate system in one coordinate system. In this paper, the spatial registration of pairwise point sets and the spatial registration of non-pairwise point sets are studied respectively. In this paper, the traditional quaternion algorithm and SVD matrix decomposition method are used to register the pairwise point set, while the nearest point iterative algorithm (Iterative Closest Point,ICP) algorithm is used for the non-pairwise point set. In order to improve the accuracy of spatial registration, this paper improves the quaternion algorithm combined with ICP algorithm for the spatial registration of pairwise point sets, and improves the ICP algorithm with Delaunay partition for the spatial registration of non-pairwise point sets. Finally, two improved algorithms are simulated. Through the analysis of the experimental results, it can be seen that the operation time of the two improved registration algorithms is longer than before, but it is still within the acceptable range. At the same time, the registration accuracy has been obviously improved, and the expected results have been achieved.
【学位授予单位】：河北工业大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：R616;TP391.41
【参考文献】
相关期刊论文 前1条
1 蒋成成;胡同森;周维;;一种改进的迭代最近点算法[J];计算机系统应用;2009年08期
，本文编号：2495248
本文链接：https://www.wllwen.com/yixuelunwen/waikelunwen/2495248.html
最近更新
教材专著
