Logo PTB

Objective

The aim of using medical imaging is to enable a reliable diagnosis. Imaging procedures which are based on the application of ionizing radiation expose the patient to radiation of a certain dose and – therefore – constitute a (minor) health risk for the patient. At present, the largest contribution to the man-induced radiation burden of the population in Germany is due to diagnosis procedures which are carried out by X-ray tomography (CT).

For this reason, the providers and users of medical imaging devices – e.g. CT or other technologies such as SPECT (single-photon emission tomography) and PET (positron-emission tomography) – are trying to optimize the relation between the image quality and the dose received by the patient. Reducing the dose, however, only makes sense as long as the image quality is good enough to reveal the information which is needed to come to a diagnosis. Both parameters – the dose as well as the image quality – are the subject of current research.

Novel tomographic reconstruction algorithms (IR, iterative reconstruction) make it possible to either reduce the dose – while maintaining, at the same time, the image quality – or to improve the image quality (with the dose staying the same). The new reconstruction procedures are, however, non-linear – whereas the parameters used so far for the quantification of image quality were based on the assumption of a linear, shift-invariant system. This no longer applies to the new procedures. For that reason, it is necessary to develop new methods to assess the image quality in an objective way. The aim is to be able to compare different imaging systems – or different recording parameters on the same system – quantitatively with each other. In this way, it will be possible to keep the radiation exposure of the patient as low as reasonably achievable.

As a consequence, the aim is to investigate first the quality of CT images. The procedures to be developed will have to be as general as possible so that they can be used for other imaging modalities as well.