Theoretical Analysis of Multi-Pinhole Brain SPECT

Nowadays, the most frequently applied setups for single-photon emission computed tomography (SPECT) of the human brain apply parallel-hole collimators. Here we investigate potential improvements in sensitivity and system resolution of brain SPECT through the use of multi-pinhole collimators with non-overlapping projections. We employ an analytical model for the geometry of a multi-pinhole setup with stationary detectors and vary (i) pinhole size, (ii) collimator-to-object distance, and (iii) detector radius, for detectors of different intrinsic resolutions (0.1, 0.5, 1, 2, 3 and 4 mm). By tuning pinhole diameters, we are able to compare the system resolution of multi-pinhole SPECT and presently used clinical devices at equal sensitivities, while a comparison of the sensitivities is performed at equal system resolutions. Multi-pinhole setups using detectors with resolutions > I mm can reach a sensitivity that is 7.4 times higher than the sensitivity of dual-head parallel-hole systems, while system resolution can be improved by a factor of 2.7. For these conventional detectors the optimal configuration has a large detector-to-collimator distance, thereby ensuring sufficient magnification of the object onto the detector to overcome the limited detector resolution. In contrast, high-resolution detectors with intrinsic resolutions < 0.3 rum should be placed very close to the collimator, resulting in a high number of de-magnified projections. In this case both system resolution and sensitivity improve considerably: for a detector resolution of 0.1 mm a 25-fold enhancement in sensitivity is achieved compared to dual-head parallel-hole devices, while the system resolution improves by a factor of 5.1. These vast improvements in performance of brain SPECT with pinholes and high-resolution detectors may open up completely new molecular imaging applications.