CWI reduces perturbations in tomography

Researcher Folkert Bleichrodt of CWI has developed new methods to improve the applicability of tomography. These methods reduce unwanted perturbations. This allows for applying tomography on a very small scale or based on a small number of measurements.

Publication date: 09-11-2015

Researcher Folkert Bleichrodt of CWI has developed new methods to improve the applicability of tomography. These methods reduce unwanted perturbations. This allows for applying tomography on a very small scale or based on a small number of measurements. He will defend his thesis on Tuesday 10 November at Leiden University.

Tomography is a technique for reconstructing the cross section of an object without physically opening it. An example is a CT scan, known from its medical application. This technique constructs a cross section of an object by determining the amount of radiation that passes through the object in various places. By making projection images from various angles it is even possible to create a 3D images from the inner structures of the object. This requires a mathematical calculation using a reconstruction algorithm.

At CWI, Bleichrodt developed a more robust version of this reconstruction algorithm. For some non-medical applications of tomography, such as imaging micro-organisms or nanomaterials, a resolution on micro or nanoscale is required. It is very challenging to create a high-quality reconstruction on such a small scale. This is caused by instabilities and perturbations during the scan, leading to inaccuracies in the calculations needed for the reconstruction. In his thesis, Bleichrodt describes mulitple methods to approach and correct unwanted movement and rotations from either the scanner or the scanned object.

Another problem in tomography is noise, introduced by scattered radiation particles or variations in the intensity of the radiation source. The effect of noise is especially significant in cases with low-intensity sources or short illumination time. To reduce this noise, Bleichrodt introduces a reconstruction algorithm that is less effected by noise. It also makes tomography applicable in cases with a small amount of projection images, such as delicate nanomaterials or living tissue, that can only be scanned a few times for safety reasons.

This research is funded by the Netherlands Organisation for Scientific Research (NWO) through the Vidi grant of Prof. Joost Batenburg.

 

Image: Wikimedia Commons