In the study of human diseases and potential treatments, small animal models play a crucial role in bridging the gap between in vitro and in vivo studies. For example, animal models of many cancers are currently available and are in widespread use. When considering imaging techniques for small animal studies, Positron Emission Tomography (PET) is well suited to the measurement of highly specific molecular interactions (e.g., glucose uptake, target-receptor binding, etc.). Small animal PET (SAP) has proven utility in the field of oncology, where it has been employed in many published studies.

Most commercially available SAP scanners employ crystal detectors and photomultiplier tubes. The price tag of crystal-based scanners ranges between $400,000 and $860,000, posing a significant barrier to the wide use of SAP in typical academic laboratories. More importantly, resolution performance of crystal cameras is fundamentally limited by detection errors, associated with the lack of measurement of the depth of interaction (DOI) of photons in the long narrow crystals required to achieve high resolution. Another significant problem unique to crystal-based SAP systems operated in 3D mode is the highly nonuniform sensitivity response over the field of view (FOV). Although commonly quoted at the center FOV, sensitivity drops linearly to zero, moving from the center to the edge of the axial field. Thus in studies requiring a whole body image, such as biodistribution studies, the mouse must be imaged much longer than would be implied by the center field sensitivity, since the animal must be repositioned or continuously scanned to cover the desired region of the body.

Proposed PET scanner using straw detectors. We propose development of a high sensitivity, 3D small animal PET system based on a revolutionary design using long linear lead-walled straw (LWS) detectors. The straw camera consists of six detector heads arranged hexagonally, as shown in the drawing, each containing thirty-five 100-straw modules (a total of 21,000 straws), and separated 22 cm from one another. Each straw is 2 mm in diameter, and 50 cm in length, providing a large axial FOV. By providing precise 3D position encoding, this approach totally eliminates DOI errors, the fundamental limiting factor preventing economical high resolution crystal based cameras. This fundamentally new approach thus provides a greatly expanded, uniform, high sensitivity FOV allowing imaging time to be reduced by as much as 4–6 fold, while at the same time providing a quantum leap in volumetric imaging resolution.

We have estimated the performance of the proposed LWS scanner in Monte Carlo simulations. The results are summarized in the table below and compared against the reported performance of the popular microPET Focus 220 scanner. For the LWS scanner, we predict a reconstructed image resolution close to 1.0 mm in all three directions, for a point source in the CFOV, degrading only very slightly to 1.1 mm at a radial offset of 10 cm. The estimated sensitivity is 6.9% in the CFOV, and the scatter fraction is 17% for a large mouse phantom, a water-filled cylinder with a diameter of 3 cm and a length of 7 cm. For a normal mouse phantom size of Ø 2.5 cm × 5.0 cm, the scatter fraction is 16%, and 19% for six phantoms inside the FOV, as illustrated in the scanner drawing above. Performance parameters promise significant improvement over the capabilities currently offered by the Focus 220, and other crystal-based scanners.





Funding

This technology has been funded by the DOE under the project name "New PET Scanner for Molecular Imaging Based on Innovative Straw Detector Technology" - DE-FG02-08ER64679 and NIH under the project name "Novel Small Animal PET Scanner Using 2 mm Lead-Walled Straw Detectors"- CA128217