Rechnergestützte Positronen-Emissions-Tomographie : ein Überblick über bisher bekannte Methoden

Since the sixties positron emission tomography (PET) has been an imaging technique in nuclear medicine for the noninvasive representation of physiological and biochemical processes in the human body. PET development was accompanied and to a large extent determined by progress in computer technology and electronics, since only digital data processing permits data collection and image reconstruction of the activity distribution in the object. The aim of the present contribution is to provide an overview of the methods currently available within the framework of computer-assisted PET from the perspective of signal and data processing. Signal detection by the PET scanner is based both on the development of photodetectors in particular as weil as on the development of scanners in general. This and the present status of research with respect to the local detection of events in the crystals show the demands made on signal processing. The field of data acquisition comprises the imaging modes (2-D/3-D), the data flow in the scanner and the possible correction steps to which the acquired data have to be subjected before reconstruction. The data reconstruction provides voxel-based volume images from the corrected acquisition data which are further processed and visualized within graphical user interfaces. The DICOM1 3.0 standard serves to integrate PET as an imaging technique in its medical environment and to standardize digital medical data. Developments in PET technology continue to Iead to a significant increase in information and data material, not least due to the consideration of dynamic processes, so that in future "intelligent" mechanisrns will be required to reduce this information to its essence