Image Quality, Accuracy of Attenuation and Scatter Corrections of the microPET Focus 120 Using the NEMA NU4-2008 Phantom

This work aimed at evaluating the image quality and the accuracy of attenuation and scatter corrections methodologies provided with the microPET Focus 120 scanner. The protocol used here is described in the forthcoming NEMA NU4-2008 document. This study included 1 h transmission experiments, with two sources (68Ge and 57Co, both one half-live old), in coincidence mode (Ge) with and without windowing, and singles mode (Ge and Co). Transmission data with Co were also collected for shorter acquisition times (515/1030/2060 s). Transmission with scatter correction, and emission with attenuation with or without scatter correction, slices were reconstructed using 2D-FBP (ramp filter). The uniformity, the recovery coefficients (RC) and the accuracy of corrections were evaluated as described in the NEMA document. We also determined the attenuation coefficient (AC) for water using the methodology applied to measure the mean pixel value for the emission slices of the uniform part of the phantom. The Co single mode measurements provided a mean AC for water closer to the theoretical value than the Ge measurements either in single or in coincidence mode even for the shortest acquisition time. Moreover, the lowest coefficients of variation per pixel of the uniform slices were measured for attenuation corrected data based on Co measurements. Also, spill-over ratio for non-emitting air and water compartments reflected a higher capability of this attenuation correction method for scatter correction. The RC measured on emission image corrected only for attenuation did not show any significant difference linked to the transmission measured methods. However, higher RC values were noted for Ge coincidence with windowing when emission data were corrected for both attenuation and scatter. Hence, Co singles transmission seems to be the most suitable method for attenuation correction on the microPET Focus 120 scanner.