A spheroid positron emission tomograph for brain imaging: a feasibility study.

Abstract

It has been long recognized that the primary advantage of imaging the brain with a positron emission tomography using GSO scintillation detectors placed on a spheroid surface is the large solid angle of acceptance for annihilation radiation, which results in improved system sensitivity and image signal-to-noise ratio. In the present study, we investigated spheroid system geometry, detector design and contribution of scattered coincidences. Scintillation detector distribution on a spheroidal surface was investigated by approximating the surface by polygons. Finding a suitable crystal for this purpose led to the development of an experimental GSO block-type detector. The fraction of scattered coincidences was experimentally evaluated using phantoms and detector pairs in conjunction with a testing platform, and the relationship between scattered fraction and phantom volume was obtained. Spheroid geometry was best implemented with a polyhedron consisting of a series of consecutive rings formed by trapezoids. An experimental block-type detector with 36 GSO scintillators and four 14-mm-diameter photomultiplier tubes, together with custom electronics, yielded a spatial resolution of 3.4 mm FWHM and an energy resolution of 18% FWHM. Using nearly "ideal" scintillation detectors with a 350-keV threshold, we found the scatter fraction to be 0.32 for a 20-cm uniform phantom, 0.22 for a 15-cm phantom and closely proportional to the square root of the phantom volume. For cerebral studies, a spheroid PET using GSO scintillators has several advantages: optimized geometry for sensitivity, a dead-time fivefold smaller than an equivalent BGO system, and appreciably better light output for improved energy resolution and detector identification. The construction of such a system is within the capabilities of present technology.