基于区域生长法的三维心血管模型的分割

发布时间：2018-04-20 17:12

  本文选题：三维重建 + 光线投射法　； 参考：《北京生物医学工程》2016年05期

【摘要】：目的医学图像能够反映人体各组织的生理或病理性的结构信息和功能信息,根据图像序列,对患者心血管组织进行三维重建,可以方便医生对血管的病灶部位做出准确诊断。本文将区域生长法(region growing method,RGM)在二维图像的基础上用于三维,并比较RGM的3种子方法对心血管模型进行三维分割的效果。方法结合VTK(Visualization Toolkit)与ITK(Insight Segmentation and Registration Toolkit)函式库,首先利用光线投射法(ray casting,RC)对患者的CT扫描图像序列进行三维重建,得到患者胸腔三维模型;之后分别利用区域生长法中连通阈值、置信连接和邻域连接3种算法进行三维分割处理,得到患者心血管模型的仿真结果。结果三维重建仿真技术,具有直观重现心血管造影序列所描述的人体胸腔部位各组织的可能性;3种分割算法的比较显示,邻域连接法相比于其他两种算法,模型的解剖结构信息丢失程度较大,血管树的分支数目及管径也明显偏小。结论使用VTK与ITK函式库,通过光线投射法的体绘制技术和区域生长法的图像分割技术,可以建立患者的心血管模型;区域生长法中的连通阈值和置信连接算法,对结果的细节保留程度最大。
[Abstract]:Objective medical images can reflect the physiological or pathological structural information and functional information of the tissues of the human body. According to the image sequence, the three-dimensional reconstruction of the cardiovascular tissue of the patient can make it convenient for doctors to make an accurate diagnosis of the location of the lesions of the blood vessels. In this paper, the region growing method (RGM) is based on the two-dimensional image. The effect of three-dimensional segmentation on the cardiovascular model was compared with the 3 seed method of RGM. Method combined with the VTK (Visualization Toolkit) and ITK (Insight Segmentation and Registration Toolkit) library, a three-dimensional reconstruction of the patient's scan image sequence was performed by using the ray projection method (ray casting) to obtain the patient's chest. Three dimensional model of the cavity is used, and then 3 algorithms are used in the region growth method, the connectivity threshold, the confidence connection and the neighborhood connection. The simulation results of the patient's cardiovascular model are obtained. Results the 3D reconstruction simulation technique has the possibility of intuitionistic reproduction of various tissues of the human thoracic cavity described by the angiocardiography sequence; 3 kinds of points. Compared with the other two algorithms, the comparison of the cut algorithm shows that the anatomical structure information of the model is much more lost than the other two algorithms, the number of branches and the diameter of the vascular tree are obviously smaller. Conclusion the heart of the patient can be established by using the VTK and ITK library, the body drawing technique of the ray projection method and the image segmentation technique of the region length method. The connectivity threshold and confidence join algorithm in the region growing method are the most important for preserving the details of the results.

【作者单位】： 上海理工大学医疗器械与食品学院;
【分类号】：R816.2;TP391.41
，

本文编号：1778677