Diémoz Paul C, Vittoria Fabio A, Hagen Charlotte K, Endrizzi Marco, Coan Paola, Bravin Alberto, Wagner Ulrich H, Rau Christoph, Robinson Ian K, Olivo Alessandro
Department of Medical Physics and Biomedical Engineering, University College London, London, UK; Research Complex at Harwell, Oxford Harwell Campus, Didcot, UK.
Department of Medical Physics and Biomedical Engineering, University College London, London, UK; Research Complex at Harwell, Oxford Harwell Campus, Didcot, UK.
Phys Med. 2016 Dec;32(12):1759-1764. doi: 10.1016/j.ejmp.2016.07.093. Epub 2016 Nov 9.
Edge illumination (EI) X-ray phase-contrast imaging (XPCI) has been under development at University College London in recent years, and has shown great potential for both laboratory and synchrotron applications. In this work, we propose a new acquisition and processing scheme. Contrary to existing retrieval methods for EI, which require as input two images acquired in different setup configurations, the proposed approach can retrieve an approximate map of the X-ray phase from a single image, thus significantly simplifying the acquisition procedure and reducing data collection times.
The retrieval method is analytically derived, based on the assumption of a quasi-homogeneous object, i.e. an object featuring a constant ratio between refractive index and absorption coefficient. The noise properties of the input and retrieved images are also theoretically analyzed under the developed formalism. The method is applied to experimental synchrotron images of a biological object.
The experimental results show that the method can provide high-quality images, where the "edge" signal typical of XPCI images is transformed to an "area" contrast that enables an easier interpretation of the sample geometry. Moreover, the retrieved images confirm that the method is highly stable against noise.
We anticipate that the developed approach will become the method of choice for a variety of applications of EI XPCI, thanks to its ability to simplify the acquisition procedure and reduce acquisitions time and dose to the sample. Future work will focus on the adaptation of the method to computed tomography and to polychromatic radiation from X-ray tubes.
近年来，伦敦大学学院一直在研发边缘照明（EI）X射线相衬成像（XPCI）技术，该技术在实验室和同步加速器应用中均显示出巨大潜力。在这项工作中，我们提出了一种新的采集和处理方案。与现有的EI检索方法不同，现有方法需要输入在不同设置配置下采集的两幅图像，而我们提出的方法可以从单幅图像中检索出X射线相位的近似图，从而显著简化采集过程并减少数据收集时间。
基于准均匀物体的假设，即折射率与吸收系数之比恒定的物体，通过解析推导得出检索方法。在已建立的形式体系下，还从理论上分析了输入图像和检索图像的噪声特性。该方法应用于生物物体的同步加速器实验图像。
实验结果表明，该方法能够提供高质量图像，其中XPCI图像典型的“边缘”信号转换为“区域”对比度，便于更轻松地解读样品几何形状。此外，检索到的图像证实该方法对噪声具有高度稳定性。
我们预计，由于该方法能够简化采集过程、减少采集时间和样品剂量，所开发的方法将成为EI XPCI各种应用的首选方法。未来的工作将集中于使该方法适用于计算机断层扫描以及X射线管的多色辐射。
