目的对现有的虚拟内窥镜成像方法进行改进 ,使之能够在PC机上运行。方法传统算法在观察点位置变化时 ,应先对所有原始体数据进行三维坐标变换后再作透视投影 ,本文算法只对被观察的那部分数据进行处理 ,减少不必要的坐标变换 ,并将坐标变换、表面检测、透视投影等紧密结合 ,在一次光线跟踪过程中完成这些计算以减小计算量 ;直接对原始体数据进行投影 ,无需先对体数据进行插值 ,因此不用保留插值的中间结果 ,减少了内存消耗 ;在成像过程中自动排除与成像结果无关的数据 ,使它们不参与插值 ,不参与投影 ,进一步提高了计算速度。结果设计出基于PC机的虚拟内窥镜系统 ,并用大小为 5 1 2× 5 1 2的CT图片 ,绘制了胸部气管的内窥图像 ,以验证本文算法的有效性。结论所提出的算法能够在PC机上实现 ,有助于虚拟内窥镜系统的推广应用 Objective To improve the existing virtual endoscopic imaging method so that it can run on a PC. Methods When the traditional algorithm changes the position of the observation point, all the original volume data should be transformed into three-dimensional coordinates and then be projected in perspective. In this paper, only the part of the data to be observed is processed to reduce unnecessary coordinate transformation, and the coordinates Transform, surface detection, perspective projection and so on, in a ray tracing process to complete these calculations to reduce the amount of computation; directly to the original volume data projection, without the need to first volume data interpolation, so do not retain the interpolation of the intermediate results, Reduce the memory consumption; in the imaging process automatically excludes the imaging data and has nothing to do, so that they do not participate in interpolation, do not participate in the projection, to further improve the calculation speed. Results A virtual endoscope system based on PC was designed and the endoscopic images of the chest trachea were drawn with CT images of 5 1 2 × 5 1 2 to verify the effectiveness of the proposed algorithm. Conclusion The proposed algorithm can be implemented on a PC, which is helpful for the popularization and application of virtual endoscope system