Simultaneous post-injection transmission and emission contamination scans in a volume imaging PET scanner

Abstract

For PET cameras without septa a transmission source of 0.5 mCi is adequate for attenuation correction while restricting near detector deadtime. However, emission contamination of post-injection transmission scans is then 30% for 1 mCi of emission activity in the FOV. The authors have developed a method to directly measure emission contamination of post-injection transmission scans using a virtual transmission source position offset by 20 degrees from the transmission source. Thus, the authors collect and subtract a masked emission contamination scan from the transmission scan. Intersection of transmission source gamma rays with the virtual source position mask is cancelled by performing the blank scan in the same manner. Thus, simultaneous and direct measurements of emission contamination are performed, removing the need for complex corrections to emission data and insensitive to activity redistributions between emission and transmission scanning. While the camera deadtime for pre-injection transmission (without emission activity in the FOV) is the same as for the blank scan, there is extra deadtime due to emission activity in the FOV during a post-injection transmission scan. This results from camera processing of the additional events from this emission activity. This excess deadtime depends primarily on the amount of emission activity. This activity is measured by detector singles countrates, so that we have generated a lookup table of extra deadtime vs countrate to correct for this. This method is accurate to 4% for 0.5 mCi in the FOV. Limitations to this method are (1) reduced transmission count densities due to deadtime which results in moderate increases in % standard deviations on measured attenuation coefficients, and (2) a transmission source angular dependent deadtime which is higher for transmission source angles that sample more than the average fraction of emission contamination events. This results in some underestimation of attenuation in high activity areas of the body. Clinical imaging using this method is now routine and often superior to pre-injection transmission since there is considerably less chance of patient motion between emission and transmission scans. From patient and phantom data taken with both pre- and post-injection transmission a lookup table of residual underestimation of attenuation correction by this method (due to angular dependent deadtimes) has been generated so that more accurate SUV calculations may be performed using post-injection transmission in a Volume Imaging PET Camera.