Radiation Physics

The Radiation Physics research group (RP, headed by Prof. L. Van Hoorebeke and Prof. M. Boone) started its research on high resolution X-ray tomography around 2006. The group is completely embedded in the Centre for X-ray tomography of the UGent (UGCT). Its research covers the complete laboratory based X-ray CT imaging chain including the study of the physical processes that govern the imaging process, simulation of the imaging chain, novel imaging techniques, design and construction of state-of-the-art very high resolution X-ray CT-systems for scientific research, tomographic reconstruction (including iterative reconstruction techniques and GPU-based code) and 3D image analysis methods.

Some of these research topics are explained in more detail below.

Design and construction of CT systems

The state-of-the-art CT systems of UGCT  are custom designed by the RP group, some of them partly in collaboration with the spin-off company X-Ray Engineering. All components are carefully selected in order to meet the needs of UGCT, and the scanners are installed in large shielding bunkers to allow for high experimental flexibility and peripheral equipment if necessary. To be able to control the scanners and peripheral equipment in an optimal way, also the scanner control software is developed in house. Through simulation, the RP group aims at obtaining a thorough understanding of the complete imaging process, in order to correct for unwanted effects such as the secondary focal spot effect in the reconstruction step. More information on the different scanners can be found on our CT scanners page.

Spectral imaging

The RP group explores the possibility to retrieve additional information using the properties of the X-ray spectrum originating from the X-ray source. This can be realized by applying dual-energy CT or K-edge imaging, or by using specific hardware to provide energy-specific information. Contrary to medical CT, the composition and size of samples scanned at our facility is very diverse, hence dual-energy CT requires sample-specific simulation in order to retrieve the optimal parameters. On the hardware side, the spectral information can be obtained from energy-dispersive detectors such as Medipix or the SLcam (or Color X-ray Camera).

Reconstruction algorithms

The CT reconstruction software Octopus, now further developed and distributed by the spin-off company XRE, was originally developed by the RP group, in a first stage for parallel-beam neutron tomography but later also for cone-beam X-ray CT.

Nowadays, the research on reconstruction software is mainly focused on iterative algebraic reconstruction techniques such as SART, and their inherent ability to correct for certain imaging artifacts by simulating the effects causing these artifacts. Examples of this are beam hardening, detector lag, etc.

Dynamic high-resolution CT

Dynamic µCT scanning or 4D-µCT is an increasingly popular tool to analyze dynamic processes within objects and structures. At UGCT, the unique gantry-based EMCT system and the datasets produced at this system pose large challenges for the reconstruction and analysis of this data. Using iterative reconstruction, post-reconstruction processing and combining Digital Volume Correlation with CT reconstruction, we exploit the potential of the full 4D-µCT dataset.

Phase contrast imaging

A common effect at high-resolution X-ray imaging is X-ray phase contrast, where the refraction of the X-rays becomes visible in a so-called mixed phase-and-amplitude image. From this mixed image, it is not possible to uniquely retrieve the local attenuation coefficient or refractive index decrement, yet several algorithms exist to approximate these values. The applicability of existing algorithms, along with the development of new algorithms, is part of the research of the RP group.

Micro-CT at Inverse Compton Scattering sources

In the scope of the Interreg project Smart*Light (web page in Dutch), the Radiation Physics research group will design and build a micro-CT end station and develop methodologies to exploit the benefits of this powerful new X-ray source.

3D analysis

In CT imaging, the output is a 3D volume of gray values, offering a huge amount of information which typically needs to be summarized into several parameters of interest. To achieve this, 3D analysis software tools are required. In order to get sufficient flexibility and custom analysis methods, the RP group develops application-tailored 3D analysis protocols in scientific collaborations. These are often based on the 3D analysis tool Morpho+, developed at the RP group.

Currently, the tool is further developed and distributed as Octopus Analysis by the spin-off company XRE.


Through collaborations with other research groups, including the Pore-Scale Processes in Geomaterials Research group and the Laboratory for Wood Technology, the two partners in UGCT, the RP group focuses on the applicability of new methods and technologies and on developing user tailored CT scanning techniques . As such, we aim at obtaining the best possible result for any given object or research question.