Georg Pelzer

On a dark-field signal generated by micrometer-sized calcifications in phase-contrast mammography

We show that a distribution of micrometer-sized calcifications in the human breast which are not visible in clinical x-ray mammography at diagnostic dose levels can produce a significant dark-field signal in a grating-based x-ray phase-contrast imaging setup with a tungsten anode x-ray tube operated at 40 kVp. A breast specimen with invasive ductal carcinoma was investigated immediately after surgery by Talbot-Lau x-ray interferometry with a design energy of 25 keV. The sample contained two tumors which were visible in ultrasound and contrast-agent enhanced MRI but invisible in clinical x-ray mammography, in specimen radiography and in the attenuation images obtained with the Talbot-Lau interferometer. One of the tumors produced significant dark-field contrast with an exposure of 0.85 mGy air-kerma. Staining of histological slices revealed sparsely distributed grains of calcium phosphate with sizes varying between 1 and 40 μm in the region of this tumor. By combining the histological investigations with an x-ray wave-field simulation we demonstrate that a corresponding distribution of grains of calcium phosphate in the form of hydroxylapatite has the ability to produce a dark-field signal which would-to a substantial degree-explain the measured dark-field image. Thus we have found the appearance of new information (compared to attenuation and differential phase images) in the dark-field image. The second tumor in the same sample did not contain a significant fraction of these very fine calcification grains and was invisible in the dark-field image. We conclude that some tumors which are invisible in x-ray absorption mammography might be detected in the x-ray dark-field image at tolerable dose levels.

Grating-based darkfield imaging of human breast tissue

Mastectomy specimens were investigated using a Talbot-Lau X-ray imaging set-up. Significant structures in the darkfield were observed, which revealed considerably higher contrast than those observed in digital mammography. Comparison with the histomorphometric image proofs that the darkfield signal correlates with a tumor region containing small calcification grains of 3 to 30μm size.

Projection angle dependence in grating-based X-ray dark-field imaging of ordered structures

Over the recent years X-ray differential phase-contrast imaging was developed for the hard X-ray regime as produced from laboratory X-ray sources. The technique uses a grating-based Talbot-Lau interferometer and was shown to yield image contrast gain, which makes it very interesting to the fields of medical imaging and non-destructive testing, respectively. In addition to X-ray attenuation contrast, the differential phase-contrast and dark-field images provide different structural information about a specimen. For the dark-field even at length scales much smaller than the spatial resolution of the imaging system. Physical interpretation of the dark-field information as present in radiographic and tomographic (CT) images requires a detailed look onto the geometric orientation between specimen and the setup. During phase-stepping the drop in intensity modulation, due to local scattering effects within the specimen is reproduced in the dark-field signal. This signal shows strong dependencies on micro-porosity and micro-fibers if these are numerous enough in the object. Since a grating-interferometer using a common unidirectional line grating is sensitive to X-ray scattering in one plane only, the dark-field image is influenced by the fiber orientations with respect to the grating bars, which can be exploited to obtain anisotropic structural information. With this contribution, we attempt to extend existing models for 2D projections to 3D data by analyzing dark-field contrast tomography of anisotropically structured materials such as carbon fiber reinforced carbon (CFRC).

Simulation framework for coherent and incoherent X-ray imaging and its application in Talbot-Lau dark-field imaging.

A simulation framework for coherent X-ray imaging, based on scalar diffraction theory, is presented. It contains a core C++ library and an additional Python interface. A workflow is presented to include contributions of inelastic scattering obtained with Monte-Carlo methods. X-ray Talbot-Lau interferometry is the primary focus of the framework. Simulations are in agreement with measurements obtained with such an interferometer. Especially, the dark-field signal of densely packed PMMA microspheres is predicted. A realistic modeling of the microsphere distribution, which is necessary for correct results, is presented. The framework can be used for both setup design and optimization but also to test and improve reconstruction methods.

Grating-based x-ray phase-contrast imaging with a multi energy-channel photon-counting pixel detector.

We have carried out grating-based x-ray differential phase-contrast measurements with a hybrid pixel detector in 16 energy channels simultaneously. A method for combining the energy resolved phase-contrast images based on energy weighting is presented. An improvement in contrast-to-noise ratio by 58.2% with respect to an emulated integrating detector could be observed in the final image. The same image quality could thus be achieved with this detector and with energy weighting at 60.0% reduced dose compared to an integrating detector. The benefit of the method depends on the object, spectrum, interferometer design and the detector efficiency.

Reconstruction method for grating-based x-ray phase-contrast images without knowledge of the grating positions

To retrieve the phase information of x-rays using a Talbot-Lau interferometer, the knowledge of the grating positions is mandatory. Transferring the interferometer technique from the laboratory to a conventional x-ray imaging system, this requirement is no longer guaranteed. This is due to distortions and vibrations which are coupled into the interferometer. Therefore, we applied a principal-component analysis to Talbot-Lau x-ray phase-contrast data. In experiments we compared this alternative approach for image reconstruction to the conventional procedure. As a result, a superior robustness of the principal-component analysis against imperfect phase-stepping data was found. Furthermore, using the proposed method, the reconstruction of x-ray phase-contrast images from randomly distributed phase-step positions is possible.

Phase-unwrapping of differential phase-contrast data using attenuation information

Phase-contrast imaging approaches suffer from a severe problem which is known in Magnetic Resonance Imaging (MRI) and Synthetic Aperture Radar (SAR) as phase-wrapping. This work focuses on an unwrapping solution for the grating based phase-contrast interferometer with X-rays. The approach delivers three types of information about the x-rayed object - the absorption, differential phase-contrast and dark-field information whereas the observed differential phase values are physically limited to the interval (-π, π]; values higher or lower than the interval borders are mapped (wrapped) back into it. In contrast to existing phase-unwrapping algorithms for MRI and SAR the presented algorithm uses the absorption image as additional information to identify and correct phase-wrapped values. The idea of the unwrapping algorithm is based on the observation that at locations with phase-wrapped values the contrast in the absorption image is high and the behavior of the gradient is similar to the real (unwrapped) phase values. This can be expressed as a cost function which has to be minimized by an integer optimizer. Applied on simulated and real datasets showed that 95.6% of phase-wraps were correctly unwrapped. Based on the results we conclude that it is possible to use the absorption information in order to identify and correct phase-wrapped values.

Optimization procedure for a Talbot-Lau x-ray phase-contrast imaging system

In Talbot-Lau x-ray imaging, the fringe visibility provided by the interferometer is a crucial quality parameter to preserve high quality images at an acceptable dose level. The noise of the obtained differential phase signal and the dark-field image is directly influenced by the visibility. To optimize the performance of such an interferometer, we use wave-field simulations to investigate the effect of the phase grating G1. Therefore, we varied the grating parameters duty cycle and grating bar height. Each set of these parameters were evaluated for different propagation distances and for multiple x-ray spectra. In this multidimensional space the interferometer configuration with the highest visibility over a wide range of energies was selected to cover a multiple possible x-ray applications. We manufactured the optimized phase grating G1, the corresponding source grating G0 and analyzer grating G2 and compare the experimental results with the expected results obtained from simulations. The presented measurements show outstanding visibilities up to 50% using a broad x-ray spectrum. These measurements are in very good agreement to the simulation results. The achieved visibility is up to two times higher than for a standard-type setup. This enhancement results in high quality images at a reasonable dose level which we exemplarily demonstrate by imaging a foreign object in a pork trotter.

Optimisation of image reconstruction for phase-contrast x-ray Talbot-Lau imaging with regard to mechanical robustness.

X-ray grating-based phase-contrast imaging opens new opportunities, inter alia, in medical imaging and non-destructive testing. Because, information about the attenuation properties and about the refractive properties of an object are gained simultaneously. Talbot-Lau imaging requires the knowledge of a reference or free-field image. The long-term stability of a Talbot-Lau interferometer is related to the time span of the validity of a measured reference image. It would be desirable to keep the validity of the reference image for a day or longer to improve feasibility of Talbot-Lau imaging. However, for example thermal and other long-term external influences result in drifting effects of the phase images. Therefore, phases are shifting over time and the reference image is not valid for long-term measurements. Thus, artifacts occur in differential phase-contrast images. We developed an algorithm to determine the differential phase-contrast image with the help of just one calibration image, which is valid for a long time-period. With the help of this algorithm, called phase-plane-fit method, it is possible to save measurement-time, as it is not necessary to take a reference image for each measurement. Additionally, transferring the interferometer technique from laboratory setups to conventional imaging systems the necessary rigidity of the system is difficult to achieve. Therefore, short-term effects like vibrations or distortions of the system lead to imperfections within the phase-stepping procedure. Consequently, artifacts occur in all three image modalities (differential phase-contrast image, attenuation image and dark-field image) of Talbot-Lau imaging. This is a problem with regard to the intended use of phase-contrast imaging for example in clinical routine or non-destructive testing. In this publication an algorithm of Vargas et al is applied and complemented to correct inaccurate phase-step positions with the help of a principal component analysis (PCA). Thus, it is possible to calculate the artifact free images. Subsequently, the whole algorithm is called PCA minimization algorithm.

Increasing the darkfield contrast-to-noise ratio using a deconvolution-based information retrieval algorithm in X-ray grating-based phase-contrast imaging

A novel information retrieval algorithm for X-ray grating-based phase-contrast imaging based on the deconvolution of the object and the reference phase stepping curve (PSC) as proposed by Modregger et al. was investigated in this paper. We applied the method for the first time on data obtained with a polychromatic spectrum and compared the results to those, received by applying the commonly used method, based on a Fourier analysis. We confirmed the expectation, that both methods deliver the same results for the absorption and the differential phase image. For the darkfield image, a mean contrast-to-noise ratio (CNR) increase by a factor of 1.17 using the new method was found. Furthermore, the dose saving potential was estimated for the deconvolution method experimentally. It is found, that for the conventional method a dose which is higher by a factor of 1.66 is needed to obtain a similar CNR value compared to the novel method. A further analysis of the data revealed, that the improvement in CNR and dose efficiency is due to the superior background noise properties of the deconvolution method, but at the cost of comparability between measurements at different applied dose values, as the mean value becomes dependent on the photon statistics used.