摘要
目的提出一种基于图形处理器(GPU)的高效实时肺部及肺部病灶区域四维可视化解决方案。方法借助开放图形类库,编程实现序列体数据的读取、显示。依据不同组织器官的不同灰度值,设置颜色查找表中灰度值的上下限来控制显示不同组织,并在确定病灶区域位置的情况下实现病灶区域的提取及局部四维可视化。结果通过GPU加速实现医学影像四维可视化技术,对肺部10个时相的数据绘制速度达到约20帧/秒,对提取后的病灶区域的绘制速度达到31帧/秒,交互绘制速度可达到22帧/秒,播放流畅,可满足临床诊断需求。结论利用基于GPU的肺部及病灶区域四维可视化技术能全方位、多角度观察肺部及病灶区域随呼吸的实时运动。
Objective To put forward a high-efficiency real-time 4D visualization technology based on graphic processing unit （GPU）, and to extract pulmonary lesions and visualize the lung. Methods By means of open graphics libraries, the read and display of sequential volume data were realized by programming. According to different gray values of different tissue, the upper and down limits in the look-up table were set to control the display of different organs. When location of lesion region was confirmed, the lesion could be extracted, and the local part was visualized through 4D visualization tech- nology. Results 4D visualization of medical image was realized by acceleration of GPU. The rendering speed for 10 time phases of the lung was up to 9,0 fps. The rendering speed for the lesion region after extraction was up to 31 fps. The inter- action speed maintained at 2g fps. It could be played smoothly and meet the needs of clinical physicians. Conclusion The extraction and 4D visualization of pulmonary lesions make omni-directional and multi-angle observation possible. Real-time pulsate of pulmonary lesions and the interactions between lesions and surrounding normal tissues can be clearly observed.
出处
《中国医学影像技术》
CSCD
北大核心
2013年第11期1901-1905,共5页
Chinese Journal of Medical Imaging Technology
基金
国家自然科学基金(61179019
61261028)
关键词
图形处理器
可视化
图像处理
计算机辅助
体绘制
肺
Graphic processing unit
Visualization
Image processing, computer-assisted
Volume rendering
Lung
作者简介
吕晓琪（1963-），男，内蒙古包头人，博士，教授。研究方向：医学图像处理的研究。E-mail：lxiaoqil83@126．com
[通讯作者]张传亭，内蒙古科技大学信息工程学院，014010。E-mail：zctzzy2008@163．com
