MicroPET Focus 120 scanner use at high-­‐count rate

Kinetic modeling of physiological processes using imaging techniques requires an accurate
measurement of the time-activity curve of the tracer in plasma, known as the arterial input function (IF). The IF can be obtained by manual blood sampling, can be derived from PET images, or continuously measured by the use of small counting systems such as beta microprobes [1]. However, some beta microprobe systems can suffering from high background counts and low sensitivity compared to PET can obligate the use of activities higher than those typical for the imaging system. In the present study, the NEMA NU4-2008 image quality (IQ) phantom [2] was used to evaluate the image quality of the microPET Focus 120 at high activity values. Attenuation correction was obtained from transmission measurement using 57Co point source. Eight emission scans of 20 minutes were performed at decreasing activity starting from 109 MBq to 3.7 MBq (total activity in the field-of-view). To study the effect of normalization in high count rate studies, several normalization scans were performed using activities ranging between 18 and 212 MBq. Images were reconstructed with all corrections using Fourier rebinning and filtered backprojection. The mean activity and the coefficients of variation of the uniform slices were measured. All high activity reconstructed images showed a detector-block-patterned artifact with an overestimation of the counts when normalization activity is higher than that used in the IQ phantom and underestimation of the counts when normalization activity is below the activity used in the IQ phantom. Using the same high activity for acquisition and normalization considerably reduces the patterned-artifact but does not eliminate it entirely. The observed artifact is due to pulse pile-up in the detectors at high count-rates. A dedicated rejection of the pulse pile-up does not appear to have been implemented for the microPET Focus 120. An alternative would be to re-calibrate the detectors with higher activity values to prevent any pile-up effect or to create an attenuation volume into which phantoms or small animals could be inserted thus decreasing the artifact. This latter option is under development.
References:
[1] G. Warnock et al, European Journal of Nuclear Medicine and Molecular Imaging Research, 1-13 (2011)
[2] NEMA Standards Publication NU4-2008. Rosslyn, VA: National Electrical Manufacturers Association; (2008).