摘要
基于开源软件开发包ITK,使用连接门限法和快速步进法分别完成对肺实质和脊柱的医学图像的自动分割,分割效果良好.通过对分割结果的对比分析,得出两种分割方法的适用对象.使用连接门限法分割CT序列图像,并输入到VTK中,实现了胸椎的三维重建.实验结果准确还原了目标组织空间位置、几何特征等信息,各脊柱块间没有粘连,能够满足医生对局部组织的观测需求.
Based on the open source software development kit ITK, the automatic segmentation of medical image of the lung essence and spinal column was completed by using the connection threshold and the fast step methods with good segmentation effect. By comparing and analyzing the segmentation results, applicable targets of two segmentation methods were discussed. The CT sequence images were segmented using the connection threshold method, and the 3 D reconstruction of the thoracic spine was achieved by inputting image into the VTK. The experiment results accurately reappear the spatial location, geometric features and other information of the target tissue, and there is no adhesion between the vertebrae, which could meet the observation needs of the local tissues.
引文
[1]陈家新.医学图像处理及三维重建技术研究[M].北京:科学出版社,2010:1-13.
[2]杨朝晖.基于ITK读写DICOM医学图像文件[J].北京生物医学工程,2014,33(5):518-523.
[3]姜红.王利,张兆臣.ITK在医学图像区域生长分割中的应用[J].中国医学装备,2008,5(11):13-17.
[4]纪晓静.基于ITK的医学图像分割系统的研究与实现[J].科技创新与应用,2015(22):12-13.
[5]苏育挺.董博.基于ITK和VTK的脊髓扩张量成像实现[J].电子测量技术,2011,34(12):54-57.
[6]周振环,王安明,王京阳,等.医学图像分割与配准(①ITK初步分册)[M].成都:电子科技大学出版社,2007:216-219.
[7]于洋.肺部CT血管分割及三维重建[D].哈尔滨:哈尔滨工业大学,2016.
[8]徐彩云.图像分割算法的研究综述[J].电脑知识与技术,2014,10(11):2637-2639.
[9]宋静,何启冷,徐庆,等.基于ITK医学分割的应用[J].成都信息工程学院学报,2015,30(5):258-461.
[10]李振华,梅雪,郭笑妍,等.模糊水平集心脏CT图像序列分割方法[J].计算机工程与设计,2015,36(11):3030-3034,3045.
[11]张晓东,罗火灵.VTK图形图像开发进阶[M].北京:机械工业出版社,2015:1-5.
[12]李爱玲.三维医学图像可视化的研究和实现[D].大连:大连理工大学,2006.