节点文献

CT图像三维重建在眼内异物定位中的应用

3D Reconstruction for CT Image in Localization of Intraocular Foreign Body

【作者】 高爽

【导师】 董心;

【作者基本信息】 吉林大学 , 生物医学工程, 2005, 硕士

【摘要】 医学图像三维重建是目前医学图像处理领域的研究热点,属于多学科交叉的研究课题,涉及到计算机图形学、图像处理、生物医学工程等多种技术,在诊断医学、手术规划及模拟仿真等方面有广泛应用。眼内异物伤是常见的眼外伤,发生率甚高,对视力的危害严重。眼内异物定位准确、直观,病人接受X 线剂量少是眼科和影像学界一直追求的目标。CT 扫描是显示眼内异物的常用方法,早期CT 机由于软件功能的限制,三维定位需加冠状扫描,未能减少病人接受X 线剂量。在获得CT横断薄层扫描图像后,再行三维重建定位,能为临床提供很直观的影像资料,有利于临床及时设计手术方案,准确的进行手术。早期CT 机由于软件功能的限制,三维定位需加冠状扫描,未能减少病人接受X 线剂量。本文以眼内异物伤患者的眼部CT 扫描图像为原始数据,通过边缘检测提取出轮廓信息,轮廓信息以数据文件形式导入CAD 造型软件(本文使用Solidworks)中,借助软件功能重建眼球三维实体,从而可以获得眼球的三维立体影像。对眼球实体的旋转、剖切等操作可以确定异物位置,并提供给临床作为异物摘出手术的参考。

【Abstract】 Intraocular foreign bodies (IOFBs) is a common ocular trauma. The occurrence rate is very high,the danger to eyesight is serious. Because the foreign body stay lasting influence to the eyeball,in principle all foreign bodies inside the eye need to diagnose early and remove in good time,while attempting to preserve vision and restore ocular architecture. The accurate localization is the important assurance that the foreign body can be remove,preserve vision after an operation. The localization is more accurate,the damage of the eyeball in operation is probably more small and the vision after operation is probably more good. CT scans is better method that shows the foreign body. After acquiring the tomography slices,carrying on the three-dimension reconstruction can offer very intuitionistic image data for clinic,help to design the operation scheme clinically in time,carry on the surgical operation accurately. 3D reconstruction for medical images is a hot subject of medical images processing, belonging to multi-disciplinary subject,involved in computer graphics and image processing in biomedicine engineering. This text uses CT’s fault scanning to be built the three-dimension model of eyeball. The difference of the CT’s machine that owing to internal the various big hospital used,and the confidentiality of the data format of CT,and this is studied using film scan method to procure CT’s fault image the incomplete export merit ability to prepare the image outline data. The edge detection by man-made recognition with recognizing two kinds of methods voluntarily. This text is developed man-madely interactive with recognizing the edge detection method voluntarily is dead against the different faults edges. SolidWorks software is based on the CAD/CAE’s /CAM/PDM’s top of a table integrated system of Windows completely,and with his complete parameterization substance model building merit ability based on the characteristic,getting the extensive use machinery in the design,the application in the finite element model building of medical use living things mechanics is inquired into in this text. The modularization substance model building of living things structures. Using the cross section portrait of CT’s scanning living things structures,and draws the marginal coordinate of outline,then the outline edge point is drawn up and for the closed contour,carries on three dimensions substances rebuilding. 1 ) the basic principle of seat firstly to choose CT’s fault scanning should be: Not to lose,and does not distort the shape form structure size information of living things structures,and the section number used is reduced to the full, and will pay attention to sampling arrives footstep,sharp and part top,part key seats such as bottom,maximum diameter place and minimal diameter place etc as well as some transition seats. 2 ) wide edge coordinate carry on the coordinate of key edge,and exists side by side for the file. 3 ) using the splice curve to draw up among the marginal data of the outline drawn up and will bring into being of contour read in SolidWorks,will each level number according to the read in,the wire frame picture of formation model. 4) The formation of physical model uses the characteristic modeling to become the physical model next life. The angle of from characteristic aspect is seen,and the characteristic that SolidWorks2001 can constitute has the drawing

  • 【网络出版投稿人】 吉林大学
  • 【网络出版年期】2005年 06期
  • 【分类号】R779.1
  • 【下载频次】175
节点文献中: