The objective of the proposed project is to improve preclinical molecular imaging capabilities using the synthetic-collimator imaging approach, which is an innovative SPECT technique in which tomographic images are reconstructed from multi-pinhole projection data collected at multiple magnifications. Preclinical imaging tasks, particularly molecular imaging of the mouse brain, demand both high spatial resolution and good sensitivity. Sub-millimeter spatial resolution is needed to resolve brain structures, while high sensitivity is needed to avoid pharmacologic effects associated with excessive ligand mass, limit radiation dose, and keep imaging times short. To address these competing demands, we will employ silicon and germanium detectors together. Each provides better spatial resolution than can be achieved with conventional scintillation cameras, making it possible to achieve a given spatial resolution at a lower pinhole magnification. Lower magnification, in turn, allows for a greater number of pinholes per unit detector area to be used. As this particular germanium detector technology has not previously been applied to preclinical SPECT, an initial task will be to optimize the signal processing and then characterize the detector performance as a gamma camera. Then a SPECT imaging system utilizing two camera heads, each with a silicon detector and a germanium detector stacked one behind the other, will be built. This design will allow for imaging to be done using photon emissions of two different energies at two different pinhole magnifications simultaneously. In addition to improving the sensitivity through the use of a unique two detector-layer approach, this method also enables tomographic reconstruction of multiplexed projection data, which further improves sensitivity by allowing a greater number of pinholes to be used. Specialized calibration procedures and iterative reconstruction algorithms will be implemented. Computational and analytical tools will be used to guide collimator design, image-acquisition strategy, and to compare experimental measurements to expected performance. Finally, the imaging capabilities of the multi- pinhole SPECT system will be demonstrated through the measurement of striatal binding in the mouse brain of a dopamine D2-receptor radioligand. These measurements will be compared to both to ex vivo measures and to analogous imaging measurements made with a commercial small-animal SPECT system. Overall, on a per detector-area basis this approach should provide a unique combination of spatial resolution and sensitivity for preclinical SPECT imaging studies.
The overall objective is to develop an imaging system offering a combination of spatial resolution and sensitivity tailored to the demands of molecular imaging of the mouse brain. This device and the associated imaging techniques will be valuable tools for the study of neurological function and dysfunction with many applications in drug development. In addition, the development and demonstration of position-sensitive germanium detectors for SPECT imaging has the potential to impact clinical nuclear medicine due to their combination of excellent spatial resolution and unparalleled energy resolution.