Comparison of different techniques for stereotactic positron emission tomography imaging

The aim of this study was to evaluate three different techniques used for stereotactic positron emission tomography (PET) image definition: (1) PET imaging with external stereotactic radioactive markers, (2) PET imaging without external stereotactic markers and subsequent coregistration with stereotactically defined imaging modality such as computed tomography (CT) or magnetic resonance imaging (MRI), (3) PET/CT imaging with utilization of external nonradioactive markers.

Special head phantom that could be fixed in the Leksell stereotactic frame was used. The phantom was filled with fluorodeoxyglucose ((18)F-FDG) in water solution at an activity concentration of 17.5 kBq/ml simulating counts from standard brain. A spherically shaped glass test vessel (inner diameter 46 mm and wall thickness 3 mm) positioned in the head phantom was filled with FDG water solution at an activity concentration of 52.5 kBq/ml corresponding to pathologic lesion during PET imaging. Leksell stereotactic MRI indicator box was filled with FDG water solution at an activity concentration of 3.1 MBq/ml. The phantom was then stereotactically investigated on PET, PET/CT, CT and MRI. Deviations between stereotactic X, Y, Z PET coordinates of the center of the spherical vessel (simulating pathological lesion) were determined in the treatment planning system according to reference image and represented inaccuracy in stereotactic PET image definition for each of three tested methods of stereotactic PET definition.

Total spatial inaccuracy for stereotactic PET image definition based on radioactive fiducials was 1.7 and 0.7 mm for 3.4- and 2.0-mm PET slices, respectively. Total spatial PET image definition inaccuracy based on PET/CT imaging and stereotactic definition using nonradioactive CT fiducials was 0.7 mm. Total spatial PET image definition inaccuracy based on coregistration was 0.5 and 0.9 mm for coregistration with MRI and CT, respectively.

All three evaluated stereotactic PET image definition techniques indicated very good accuracy in this phantom study entirely accepted by clinical requirements for functional imaging. The most convenient stereotactic PET image definition technique seemed to be PET image coregistration either on CT or MRI. In this situation, PET imaging can be done independently on frame application (for example few days before stereotactic frame application or even in a different centre) and then coregistered with stereotactically performed CT or MRI during the stereotactic procedure. However, detailed patient study has to be performed to test image coregistration inaccuracy on real clinical data.