Department of Radiology and Radiological Sciences, Vanderbilt University Medical Center, Nashville, Tennessee 37232, USA.
Med Phys. 2013 Aug;40(8):081911. doi: 10.1118/1.4813295.
The purpose of this work is to (1) demonstrate laboratory measurements of phase shift images derived from in-line phase-contrast radiographs using the attenuation-partition based algorithm (APBA) of Yan et al. [Opt. Express 18(15), 16074-16089 (2010)], (2) verify that the APBA reconstructed images obey the linearity principle, and (3) reconstruct tomosynthesis phase shift images from a collection of angularly sampled planar phase shift images.
An unmodified, commercially available cabinet x-ray system (Faxitron LX-60) was used in this experiment. This system contains a tungsten anode x-ray tube with a nominal focal spot size of 10 μm. The digital detector uses CsI∕CMOS with a pixel size of 50×50 μm. The phantoms used consisted of one acrylic plate, two polystyrene plates, and a habanero pepper. Tomosynthesis images were reconstructed from 51 images acquired over a ±25° arc. All phase shift images were reconstructed using the APBA.
Image contrast derived from the planar phase shift image of an acrylic plate of uniform thickness exceeded the contrast of the traditional attenuation image by an approximate factor of two. Comparison of the planar phase shift images from a single, uniform thickness polystyrene plate with two polystyrene plates demonstrated an approximate linearity of the estimated phase shift with plate thickness (-1600 rad vs -2970 rad). Tomographic phase shift images of the habanero pepper exhibited acceptable spatial resolution and contrast comparable to the corresponding attenuation image.
This work demonstrated the feasibility of laboratory-based phase shift tomosynthesis and suggests that phase shift imaging could potentially provide a new imaging biomarker. Further investigation will be needed to determine if phase shift contrast will be able to provide new tissue contrast information or improved clinical performance.
本工作旨在:(1)展示基于衰减分区的 Yan 等人提出的算法(APBA)从线相位对比射线照片得出的相位偏移图像的实验室测量结果;(2)验证 APBA 重建图像是否符合线性原理;(3)从一系列角度采样的平面相位偏移图像重建断层合成相位偏移图像。
本实验使用了一台未经修改的、市售的厢式 X 射线系统(Faxitron LX-60)。该系统包含一个钨阳极 X 射线管,标称焦点尺寸为 10μm。数字探测器使用 CsI∕CMOS,像素尺寸为 50×50μm。使用的体模包括一个亚克力板、两个聚苯乙烯板和一个墨西哥胡椒。断层合成图像是从 51 张在 ±25°弧线上采集的图像重建的。所有相位偏移图像均使用 APBA 重建。
来自具有均匀厚度的亚克力板的平面相位偏移图像得出的图像对比度比传统衰减图像的对比度大约高出两倍。单个、均匀厚度的聚苯乙烯板的平面相位偏移图像与两个聚苯乙烯板的平面相位偏移图像的比较表明,估计的相位偏移与板厚度之间具有近似的线性关系(-1600 rad 与-2970 rad)。墨西哥胡椒的层析相位偏移图像表现出可接受的空间分辨率和与相应衰减图像相当的对比度。
本工作证明了基于实验室的相位偏移断层合成的可行性,并表明相位偏移成像有可能提供新的成像生物标志物。需要进一步研究以确定相位偏移对比度是否能够提供新的组织对比度信息或改善临床性能。