The YAP-(S)PET scanner

The use of positron emission tomography (PET) and single-photon emission computed tomography (SPECT) in small animal studies has undergone a significant increase in recent years as a consequence of the development of dedicated scanners. Clinical scanners do not provide the necessary requirements for small animal investigations, in particular, lacking high resolution and high sensitivity.
One of the lastest achievements in this area is the YAP-(S)PET scanner. The scanner was originally developed at the Departments of Physics of the University of Ferrara and University of Pisa, Italy, and is now commercially available.
The YAP-(S)PET is the only scanner that combines the PET and SPECT techniques and offers the unique possibility of developing new and interesting protocols for the investigation of many biological phenomena more effectively than with PET or SPECT modalities alone.
The scanner is made up of four detector heads, each one composed of a 4 x 4 cm²of Yttrium Aluminum Perovskite activated by Cerium(YAlO₃:Ce or YAP:Ce) matrix of 20 x 20 crystals, 2 x 2 mm wide and 25 deep. Each finger crystal is optically insulated from adjacent ones by a 5 μm reflecting layer.
The matrix is directly coupled to a position sensitive photomultiplier (PS-PMT, Hamamatsu R2486) with 3 mm in diameter and 0.5 mm of intrinsic spatial resolution. The four modules are positioned on a rotating gantry (Fig.1) and the opposite detectors are in time coincidence when used in PET mode. In PET mode the acquisition is 3D. Switching to SPECT modality can be done by replacing, for each head, the tungsten septum (used in PET for shielding the detector from the background outside the field of view) with a high-resolution parallel-hole, lead collimator (0.6 mm Ø, 0.15 mm thick). For both PET and SPECT modalities the scanner has an axial field of view of 4 cm and a transaxial FOV diameter of 4 cm. The system can operate in 2D and 3D data acquisition mode.
Data acquisition, data analysis and image reconstruction are performed by a user-friendly graphic interface. Both FBP (Filtered Back Projection) and EM (Expectation Maximization) algorithms can be used for image reconstruction.

Figure 1: Photograph of the YAP-(S)PET scanner (left) and zoom of the four rotating heads (right).

The scanner performance

The YAP-(S)PET scanner peformace have been evaluated both on phantoms and animals. Table 1 reports the main YAP-(S)PET performance characteristics for both PET and SPECT modality.

PET		SPECT
Sensitivity @ CFOV (50-850 keV)	1.87 % (18.7 cps/kBq)	Sensitivity (140-250 keV) (costant over the FOV)	0.03 <89> (30 cps/MBq)
Scatter fraction	17.6 %
Peak NEC rate	23.5 kcps @ 31.5 MBq
Spatial resolution (FWHM) @ CFOV (EM algorithm)		Spatial resolution (FWHM) (costant over the FOV) (EM algorithm)
Radial	1.52 mm	Radial	3.1 mm
Tangential	1.62 mm	Tangential	2.9 mm
Axial	2.10 mm	Axial	3.3 mm

Table 1: YAP-(S)PET performance in PET and SPECT modality.

Animal and phantom studies

The imaging performance were evaluated using a Derenzo-like phantom. The sizes of the Derenzo-like rods are 3.0, 2.5, 2.0, and 1.5 mm Ø. The centre-to-centre distance between adjacent rods is twice the rod diameter. Figure 2 shows a phantom photo (center) and the central slice of the PET and SPECT acquisition (left and right respectively). In PET mode the system can resolve the 1.5 mm hot rods, while in SPECT modality only the 2.5 mm rods are clearly resolved.

Figure 2:(b) Photograph of the Derenzo phantom. Starting from the right and going anti-clock wise we have: 3.0, 2.5, 2.0 and 1 mm diameter holes. (a) PET image of the Derenzo phantom obtained for an energy window of 50-850 keV and using EM algorithm. (c) Reconstructed image of the Derenzo phantom in SPECT mode: energy window of 140-250 keV, EM algorithm.

Figure 3 shows the glucose metabolism of a normal rat obtained injecting 37 MBq of ¹⁸F-FDG. Cerebral activity can be clearly distinguished from extracerebral activity. More specifically, activity in the Harderian glands is clearly distinct from brain activity. Within the brain, principal brain structures such as cortex, thalamus and striatum are well resolved.

Figure 3: Transaxial (coronal) section (0.5 x 0.5 x 2 mm voxel size), of a rat brain injected with 37 MBq of ¹⁸F-FDG and acquired for 45 minutes, after 30 minutes of uptake. The images are arranged from rostral (upper left) to caudal (lower right). The right hemisphere of the animal is represented on the right hand side of the figure, and the dorsal aspect corresponds to the top of the figure. EM algorithm with 30 iterations was used for image reconstruction.

Figure 4 reports some images of the studies perfomed with the the high-affinity ligand for the postsynaptic D2-like dopamine receptor, ¹⁸F-Fallypride. The studies were applied to two groups of healthy male rats (CD, 300 g). One group of animals was treated with an intraperitoneal injection of 50 mg/(kg body weight) amisulpride in order to block the binding of the ¹⁸F-Fallypride; the other one was not treated. All the animals were anesthetized with Chloralhydrate 7 %, injected via a lateral tail vein with 37 MBq of ¹⁸F-Fallypride and immediately scanned with a dynamic protocol.

Figure 4: Transaxial (left) and coronal (right) slices of a healthy male rat (upper figure) and of a male rat treated with amisulpride in order to block binding of ¹⁸F-Fallypride (lower figure). All the animals were injected with 37 MBq of ¹⁸F-Fallypride and immediately scanned. Voxel size is 0.5 mm × 0.5 mm in the transaxial plane and 2 mm in the axial direction.The images were reconstructed using EM algorithm with 40 iterations.

The rat miocardium perfusion studies showed in Figure 5 was performed in SPECT with ^99mTc Myoview. The rat (204 g) was injected with 300 MBq of ^99mTc Myoview and acquired for 80 min. after and uptake of 180 min. The EM algorithm was used.

Figure 5: Axial (left) and coronal sections (right) of rat miocardium perfusion studies performed in SPECT with ^99mTc Myoview.

Literature

A.Bartoli, N.Belcari, D.Stark, S.Hohnemann, M.Piel, M.Jennerwein, U.Schmitt, J.Tillmanns, O.Thews, C.Hiemke, F.Roesch, A.Del Guerra, " Preliminary Assessment of the imaging capabilities of the YAP-(S)PET small animal scanner in neuroscience", Nuclear Instr and Methods in Phys Res A 2007, 569(2), 488-491.

A.Del Guerra, A.Bartoli, N.Belcari, D.Herbert, A.Motta, A.Vaiano, G.Di Domenico, N.Sabba, E.Moretti, G.Zavattini, M.Lazzarotti, L.Sensi, M.Larobina, "Performance Evaluation of the fully engineered YAP-(S)PET Scanner for Small Animal Imaging", IEEE Trans. Nucl. Sci. (2006) Vol. 53(3), 1078-1083.

A. Motta, C. Damiani, A. Del Guerra, G. Di Domenico, G. Zavattini, "Use of a fast deconvolution EM algorithm for 3-D imaging reconstruction with the YAP-PET tomograph", Comp. Med. Im. and Graph 26 293-302 (2002).

A. Del Guerra, C. Damiani, G. Di Domenico, A. Motta, L. Sartori, G. Zavattini, "An integrated PET-SPECT small animal imager: preliminary results", IEEE Trans Nucl Sci 47 1537-1540 (2000).

A. Del Guerra, Di Domenico, M. Scandola, G. Zavattini, "High spatial resolution small animal YAPPET", Nucl. Instr. and Meth. Phys. Res. A409 537-541 (1998).

I.S.E. Ingegneria dei Sistemi Elettronici s.r.l., Vecchiano, Pisa, Italy, Web Page.