Skip to Main Content
The livetime of several PET systems has been characterized in terms of a paralysing time after each coincidence event. If part of the deadtime is due to light collection and position encoding in a crystal array, the system deadtime is a function of the singles-to-coincidence-count ratio which in turn depends on the object size and radioisotope distribution. In this case it is necessary to make an independent measurement of the detector livetime. Since both detectors must be "live" to record a coincidence event, the detector contribution to system livetime is more pronounced in tomographs using position encoding schemes. Since "dead" detectors cannot contribute random events, the randoms rate will be less if deadtime is caused by the detectors rather than by the coincidence circuit. After analysing the sources of lost events at each stage in the encoding process we propose that, rather than using empi rical estimates, hardware be used to compute these losses.