Abstract
In edge-illumination x-ray tomography, which yields phase contrast in addition to attenuation contrast, spatial resolution is decoupled from the size of the x-ray source's focal spot, and of the detector pixels, as the primary x-ray beam is structured in an array of narrow beamlets. This study provides a theoretical framework to explain how beamlet width sets spatial resolution in a tomographic phase-contrast image, and examines the effect of sampling during tomographic scans. The results will inform the design of advanced experimental setups and acquisition schemes, and help us understand how resolution is affected by constraints on scan time or dose.
Highlights
Edge-illumination tomography [1,2] is a technique for 3D, phase-sensitive x-ray imaging, which has been under continuous development alongside other approaches for this purpose, such as propagation-based imaging [3], analyzer-based imaging [4], grating interferometry [5], and lately speckle-based techniques [6,7,8]
The option to choose between step-and-shoot and continuous dithering, and thereby to manipulate the accessible frequency content, implies that edge-illumination x-ray tomography exhibits some degree of flexibility in terms of spatial resolution, with resolutions ranging from p down to w accessible
We present a theoretical framework for spatial resolution in edge-illumination x-ray tomography, a metric that has specific features as it is decoupled from the source and pixel dimensions due to the structuring of the primary beam into an array of beamlets
Summary
Edge-illumination tomography [1,2] is a technique for 3D, phase-sensitive x-ray imaging, which has been under continuous development alongside other approaches for this purpose, such as propagation-based imaging [3], analyzer-based imaging [4], grating interferometry [5], and lately speckle-based techniques [6,7,8]. Besides providing access to phase-contrast, an interesting side effect of using beamlets is that it decouples the spatial resolution from the dimensions of the detector pixels and the source, which are the driving factors for Their work provides a simple but sufficiently accurate description of how a tomographic imaging system responds in frequency space to an impulse it receives in real space; it has been demonstrated that the spectral content of the Radon transform is confined to a well-defined area that has the shape of a bowtie This has proven to be a powerful insight, for characterizing the spatial resolution in tomographic scanners in general, and for, e.g., improving the image reconstruction process [14] and developing new approaches to region-of-interest scanning [15]. This article concludes with a brief summary and a critical evaluation of the work
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call