医学图像三维表面模型重建

Surface shading of three-dimensional medical objects

三维表面模型重建是医学图像处理的重要内容，是将ＣＴ和ＭＲＩ等医疗影像设备获得的二维图像重建成三维立体表面显示图像的过程．本文重点研究了其重建算法的具体实现：（１）在建模上采用了基于体素的三维物体体积重建；（２）在显示上提出了基于Ｒａｙ－Ｃａｓｔｉｎｇ的三维物体二维直接显示技术．所以，和传统采用三角面拟合的方法相比，具有图像重建过程无需生成中间数据、无需进行３Ｄ物体的边界检测等优点，同时该算法并行性好，易于采用硬件方法实现加速．因此，本研究为医学图像三维表面模型重建应用软件的开发打下了理论基础．

In many medical imaging modalities, such as CT, MR or PET, a 3D anatomy is usually represented by a set of planar images. For its visualization, we use a voxel model with a ray casting algorithm. Unlike the triangle method, this algorithm does not require a detection of the boundary of the 3D object beforehand, but derives the surface image display directly from the object for each viewing angle. Other attractive features of the algorithm include the capability of performing ray castings for different screen pixels concurrently. Hardware implementation to speed up the surface rendering may be facilitated by taking advantage of this concurrent capability.