PET Phantom Imaging Protocol for 90Y Quantification

Data were acquired on a variety of PET scanners (Table 1) from the major vendors, with a number of different reconstructions from systems equipped both with and without time of flight (ToF) and resolution recovery (RR). Each site followed an identical experimental protocol utilizing the NEMA 2007/IEC 2008 PET Body Phantom (Data Spectrum Corporation, NC), with a volume of about 10 L containing a “cold” (nonradioactive) solid insert (diameter 51 mm) and six fillable spheres of various diameters (∅ 10, 13, 17, 22, 28 and 37 mm) filled to an approximate eight-to-one sphere-to-background ratio with ⁹⁰Y-chloride (YCl₃) provided in a constant specific activity (PerkinElmer, Waltham, MA).
Table 1

The scanners contributing data to the study according to vendor and model (all scanners equipped with standard reconstruction corrections for attenuation, scatter and random events)

Vendor	Model	Crystal material	Additional corrections	Number of scanners	Number of reconstructions
GE Healthcare	Discovery 690, 710	LYSO	ToF, RR	9	21
	Discovery 600, Discovery ST (E)	BGO	With or without RR	9	16
	Discovery RX	LYSO	–	3	7
Philips	Gemini TF	LYSO	ToF, RR	9	9
Siemens	Biograph mCT	LSO	ToF, RR	19a	28
	Biograph (various)	LSO	With or without RR	19	28

ToF’ time-of-flight, RR’ resolution recovery (point spread function recovery)

^aIncluding two systems with the new continuous bed motion technology

Each site was required to measure the phantom volume and the delivered ⁹⁰Y solution in the departmental dose calibrator for comparison with the shipping certificate. The entire delivery vial was added to a volume of 1,300 ml, and this solution was used to fill the phantom spheres, before the reminder of solution was added to the background compartment, with the addition of EDTA/DTPA to the contents to prevent the YCl₃ sticking to the phantom walls. This allowed an eight-to-one sphere-to-background ratio, in keeping with the NEMA NU 2-2007 [24 ] image quality guidelines, and was thought to require minimal phantom manipulation and activity handling at sites. Residual in the vial was estimated through re-measuring the vial in the dose calibrator after reconstitution to the initial volume with water. Residual in the needle and syringe was taken as negligible.