Le. Adam et al., PERFORMANCE EVALUATION OF THE WHOLE-BODY PET SCANNER ECAT EXACT HR+ FOLLOWING THE IEC STANDARD, IEEE transactions on nuclear science, 44(3), 1997, pp. 1172-1179
The performance parameters of the whole-body PET scanner ECAT EXACT HR
+ (CTI/Siemens, Knoxville, TN) were determined following the standard
proposed by the International Electrotechnical Commission (IEC). The t
ests were expanded by some measurements concerning the accuracy of the
correction algorithms and the geometric fidelity of the reconstructed
images. The scanner consists of 32 rings, each with 576 BGO detectors
(4.05 x 4.39 x 30 mm(3)), covering an axial field-of-view of 15.5 cm
and a patient port of 56.2 cm. The transaxial FWHM determined by a Gau
ssian fit in the 2D (3D) mode is 4.5 (4.3) mm at the center. It increa
ses to 8.9 (8.3) mm radially and to 5.8 (5.2) mm tangentially at a rad
ial distance of r = 20 cm. The average axial resolution varies between
4.9 (4.1) mm FWHM at the center and 8.8 (8.1) mm at r = 20 cm. The sy
stem sensitivity for unscattered true events is 5.85 (26.4) cps/Bq/ml
(measured with a 20 cm cylinder). The 50% dead-time losses were reache
d for a true event count rate (including scatter) of 286 (500) kcps at
an activity concentration of 74 (25) kBq/ml. The system scatter fract
ion is 0.24 (0.35). With the exception of the 3D attenuation correctio
n algorithm, all correction algorithms work reliably. The results reve
al that the ECAT EXACT HR+ has a good and nearly isotropic spatial res
olution. Due to the small detector elements, however, it has a low sli
ce sensitivity which is a limiting factor for image quality.