In-line phase-contrast X-ray imaging provides images where both absorption and refraction contribute. For quantitative analysis of these images, the phase needs to be retrieved numerically. There are many phase-retrieval methods available. Those suitable for phase-contrast tomography, i.e., non-iterative phase-retrieval methods that use only one image at each projection angle, all follow the same pattern though derived in different ways. We outline this pattern and use it to compare the methods to each other, considering only phase-retrieval performance and not the additional effects of tomographic reconstruction. We also outline derivations, approximations and assumptions, and show which methods are similar or identical and how they relate to each other. A simple scheme for choosing reconstruction method is presented, and numerical phase-retrieval performed for all methods.

We report on the observation and application of near-field speckles with a laboratory x-ray source. The detection of speckles is possible thanks to the enhanced brilliance properties of the used liquid-metal-jet source, and opens the way to a range of new applications in laboratory-based coherent x-ray imaging. Here, we use the speckle pattern for multimodal imaging of demonstrator objects. Moreover, we introduce algorithms for phase and dark-field imaging using speckle tracking, and we show that they yield superior results with respect to existing methods.

The pyramid wavefront sensor is very similar to the Fourier knife-edge test, but employs dynamic modulation to quantify the phase derivative. For circular modulation, we compare approximate geometrical optics calculations, more exact diffraction calculations, and experimental results. We show that both the sinusoidal and the approximate linear relationship between wavefront derivative and wavefront sensor response can be derived rigorously from diffraction theory. We also show that geometrical, diffraction and experimental results are very similar, and conclude that the approximate geometrical predictions can be used in place of the more complex diffraction results.

We present a high-brightness 24-keV electron-impact microfocus x-ray source based on continuous operation of a heated liquid-indium/gallium-jet anode. The 30-70 W electron beam is magnetically focused onto the jet, producing a circular 7-13 μm full width half maximum x-ray spot. The measured spectral brightness at the 24.2 keV In K(α) line is 3 × 10(9) photons∕(s × mm(2) × mrad(2) × 0.1% BW) at 30 W electron-beam power. The high photon energy compared to existing liquid-metal-jet sources increases the penetration depth and allows imaging of thicker samples. The applicability of the source in the biomedical field is demonstrated by high-resolution imaging of a mammography phantom and a phase-contrast angiography phantom.

This is the first application of liquid-metal-jet x-ray sources for whole-body small-animal x-ray imaging. In absorption, the method allows high-resolution tomographic skeletal imaging with potential for significantly shorter exposure times due to the power scalability of liquid-metal-jet sources. In phase contrast, the authors use a simple in-line arrangement to show distinct tumor demarcation of few-mm-sized tumors. This is, to their knowledge, the first small-animal tumor visualization with a laboratory phase-contrast system.