小动物锥束CT分角度几何校正

目的 改善小动物锥束CT(CBCT)系统的空间分辨力。 方法 对由3个钢制小球组成的校正模型进行360°扫描,根据质心法计算小球中心的投影位置,然后计算光源的位置、旋转轴偏移等7个几何参数,最后对在旋转扫描过程中由于系统的不稳定而发生变化的部分几何参数进行分角度计算,得出投影角度相关的几何校正参数。在对被测物体进行扫描后,利用上述几何参数和分角度校正参数,采用FDK算法进行图像重建。 结果 经过分角度几何校正后,重建图像的空间分辨力提高了2 lp/mm。 结论 分角度几何校正方法能够很好地解决系统不稳定造成几何参数在扫描中发生轻微变化的问题,显著提高CBCT的空间分辨力。

Projection angle based geometric correction for micro cone-beam CT

Objective To improve the spatial resolution of cone-beam CT (CBCT) for small animal. Methods Firstly, 360° scan of a three-ball phantom was performed. Then, 7 geometric parameters such as the position of the X-ray source, shift of the axis of rotation, etc. The projection of the center of the balls were calculated by centroid estimation. Finally, the geometric parameters were determined at each angle for the instability of the system during the rotation. Image reconstruction was done by FDK algorithm with projection angle depended on geometric parameters. Results The resolution of the reconstructed image was improved for about 2 lp/mm. Conclusion Projection angle based geometric correction can solve the problem of small variance of the geometric parameters during scanning, and significantly improve the spatial resolution of CBCT.