Quantitative comparison of 3D and 2D PET with brain studies

The statistical superiority of 3D PET compared to 2D PET for brain neuroreceptor and 18-F-fluorodeoxyglucose (FDG) scanning is accepted. However, further validation of the quantitative aspects of data acquired in 3D mode is necessary. A 3D data processing protocol that involves iterative convolution subtraction scatter correction, detector normalization including axial and radial geometric factors, attenuation correction extracted from a 2D transmission scan, the Kinahan-Rogers reconstruction algorithm and region-of-interest based calibration factors was tested on 3D data from eleven human subjects using /sup 11/Cd-hydrotetrabenazine, /sup 11/C-Schering 23390 and FDG as tracers, 2D scans were performed on the same subjects and the DV, DVR and LMRGlu values obtained from these were used as reference. The comparison showed good quantitative agreement between the two acquisition modes.