Scatter kernel estimation with an edge-spread function method for cone-beam computed tomography imaging
Scatter kernel estimation with an edge-spread function method for cone-beam computed tomography imaging
复制标题
DOI：
10.1088/0031-9155/53/23/006
复制
发表时间：
2008-12-07
影响因子：
3.5
通讯作者：
Zhu, X. Ronald
中科院分区：
文献类型：
--
作者：
Li, Heng;Mohan, Radhe;Zhu, X. Ronald
千伏X射线锥束计算机断层扫描(CBCT)的临床应用受到了CBCT图像质量的限制，这通常是由于投影数据中的大量散射分量造成的。本文描述了一种推导CBCT成像系统散射核的实验方法。估计的散射核可用于去除CBCT投影图像中的散射分量，从而提高重建图像的质量。散射辐射被近似为与深度相关的铅笔束核，并使用边缘扩展函数(ESF)方法进行推导。ESF的几何形状是通过将3 mm厚的铅片放置在一堆板状固体水模体上创建的半波束块来实现的。在铅板(半封闭)和未封闭(未封闭)的情况下，测量了10个水当量厚度(湿)，范围从0 cm到41 cm，然后在不假设任何经验试函数的情况下推导出相应的铅笔束散射核或点扩散函数(PSF)。用体模实验验证了推导的散射核函数的正确性。然后在重建过程中加入散射校正，以提高图像质量。对于直径为32 cm的圆柱体模型，重建图像的平坦度从22%提高到5%。将该方法应用于接受骨盆和肺部影像引导治疗的患者的CBCT图像时，选定感兴趣区(ROI)的变化从>；300HU减少到<；100HU。结果表明，利用散射核的散射抑制技术有效地抑制了CBCT中由散射引起的伪影。
The clinical applications of kilovoltage x-ray cone- beam computed tomography (CBCT) have been compromised by the limited quality of CBCT images, which typically is due to a substantial scatter component in the projection data. In this paper, we describe an experimental method of deriving the scatter kernel of a CBCT imaging system. The estimated scatter kernel can be used to remove the scatter component from the CBCT projection images, thus improving the quality of the reconstructed image. The scattered radiation was approximated as depth-dependent, pencil-beam kernels, which were derived using an edge-spread function (ESF) method. The ESF geometry was achieved with a half-beam block created by a 3 mm thick lead sheet placed on a stack of slab solid-water phantoms. Measurements for ten water-equivalent thicknesses (WET) ranging from 0 cm to 41 cm were taken with (half-blocked) and without (unblocked) the lead sheet, and corresponding pencil-beam scatter kernels or point-spread functions (PSFs) were then derived without assuming any empirical trial function. The derived scatter kernels were verified with phantom studies. Scatter correction was then incorporated into the reconstruction process to improve image quality. For a 32 cm diameter cylinder phantom, the flatness of the reconstructed image was improved from 22% to 5%. When the method was applied to CBCT images for patients undergoing image-guided therapy of the pelvis and lung, the variation in selected regions of interest (ROIs) was reduced from > 300 HU to < 100 HU. We conclude that the scatter reduction technique utilizing the scatter kernel effectively suppresses the artifact caused by scatter in CBCT.