双源CT心脏图像中二尖瓣的分割及三维重建

目的 对双源CT(DSCT)图像中心脏二尖瓣进行分割和三维重建,为二尖瓣结构和功能异常分析提供参考。 方法 采用两步分割法对DSCT图像中二尖瓣分割:首先利用基于区域竞争主动轮廓模型的快速水平集算法(RCAC-FLSA)对经过双边滤波处理后图像进行初步分割,得到心脏对比剂增强区域;然后在灰度拉伸处理的基础上,结合ROI,再次利用RCAC-FLSA对上一步分割结果进行分割,得到心脏二尖瓣区域;最后对二尖瓣进行恢复。在Visual C++ 2005平台上结合OpenGL开发三维重建与显示平台,利用基于三维纹理映射的体绘制方法进行三维重建,并且加入伪彩色处理和透明度处理,以增强三维重建的立体效果。 结果 成功分割出一系列DSCT心脏图像中的二尖瓣,结合伪彩色处理和透明度处理的三维重建与显示平台,可获得二尖瓣的逼真重建。 结论 两步分割算法能有效分割DSCT心脏图像中的二尖瓣;结合伪彩色处理和透明度处理的三维重建与显示平台,能提供逼真的三维重建效果。

Segmentation and 3D reconstruction of mitral valve in dual source CT images

Objective To get the segmentation of mitral valve from dual source CT (DSCT) images and reconstruct the mitral valve, in order to provide references for analysis of abnormal structure and function of mitral valve. Methods A two-step-segmentation algorithm was designed to get the segmentation of the mitral valve. Firstly, region competition active contour modal based on fast level-set algorithm (RCAC-FLSA) was applied to extract the region of heart that enhanced by contrast media after bilateral filter process. Secondly, following the process of gray level extension, RCAC-FLSA was used again to extract the mitral valve region with the limitation of region-of-interest. Lastly, the mitral valve was recovered. With the result of segmentation, a 3D reconstruction platform was implemented with OpenGL in the environment of Visual C++ 2005. Pseudo-color and transparency process were used to achieve realistic 3D reconstruction effects. Results With these methods, the mitral valve in a series of DSCT images were segmented successfully. Realistic 3D reconstruction of the mitral valve was achieved. Conclusion Using two-step-segment algorithm can get accurate segmentation of the mitral valve in DSCT images, whereas 3D reconstruct platform can provide realistic 3D view of the mitral valve.