No abstract

No abstract

No abstract

A biometrical analysis of the dinoflagellate cyst Lingulodinium machaerophorum (Deflandre and Cookson 1955) Wall, 1967 in 144 globally distributed surface sediment samples revealed that the average process length is related to summer salinity and temperature at a water depth of 30 m by the equation (salinity/temperature) = (0.078*average process length + 0.534) with R² = 0.69. This relationship can be used to reconstruct palaeosalinities, albeit with caution. The particular ecological window can be associated with known distributions of the corresponding motile stage Lingulodinium polyedrum (Stein) Dodge, 1989. Confocal laser microscopy showed that the average process length is positively related to the average distance between process bases (R²=0.78), and negatively related to the number of processes (R²=0.65). These results document the existence of two end members in cyst formation: one with many short, densely distributed processes and one with a few, long, widely spaced processes, which can be respectively related to low and high salinity/temperature ratios. Obstruction during formation of the cysts causes anomalous distributions of the processes. From a biological perspective, processes function to facilitate sinking of the cysts through clustering.

For years X-ray computerized tomography has been widely used as a medical diagnostical tool.This non-destructive technique soon turned out to be very practical in rock material research. In the 1970s CT was introduced in material research while in the nineties, micro-CT became an important non-destructive research technique. Recently nano -CT is being developed creating even more possibilities for the 3D visualization of small objects. In this paper CT, micro-CT and nano-CT are specified and discussed. Several applications illustrate the possibilities, specific advantages and limitations of each machine. Like every technique some restrictions occur, but X-ray CT turns out to be an emerging non-destructive analytical technique with a lot of possibilities in material research.

Three-dimensional (3D) analysis is an essential tool to obtain quantitative results from 3D datasets. Considerable progress has been made in 3D imaging techniques, resulting in a growing need for more flexible, complete analysis packages containing advanced algorithms. At the Centre for X-ray Tomography of the Ghent University (UGCT), research is being done on the improvement of both hardware and software for high-resolution X-ray computed tomography (CT). UGCT collaborates with research groups from different disciplines, each having specific needs. To meet these requirements the analysis software package, Morpho+, was developed in-house. Morpho+ contains an extensive set of high-performance 3D operations to obtain object segmentation, separation, and parameterization (orientation, maximum opening, equivalent diameter, sphericity, connectivity, etc.), or to extract a 3D geometrical representation (surface mesh or skeleton) for further modeling. These algorithms have a relatively short processing time when analyzing large datasets. Additionally, Morpho+ is equipped with an interactive and intuitive user interface in which the results are visualized. The package allows scientists from various fields to obtain the necessary quantitative results when applying high-resolution X-ray CT as a research tool to the nondestructive investigation of the microstructure of materials.

When a very low absorbing sample is scanned at an X-ray CT-setup with a microfocus X-ray tube and a high resolution detector, the obtained projection images not only contain absorption contrast, but also phase contrast. While images without a phase signal can be reconstructed very well, such mixed phase and absorption images give rise to severe artefacts in the reconstructed slices. In this paper, a method is described that applies a correction to these mixed projections, in order to remove the phase signal. These corrected images can then be processed using a standard filtered back projection algorithm to obtain reconstructions with only little or no phase artefacts. This new method, which we call the Bronnikov Aided Correction (BAC), can be used in a broad variety of applications and without much additional effort. It is tested on a biological and a pharmaceutical sample, results are evaluated and discussed by comparing them with conventional reconstruction methods.

No abstract