A Compton Camera Resolution Enhancement by Increasing the Number of Sensors per Readout Channel

Compton cameras (CCs) detectors are widely used for gamma-ray source imaging. Various applications employing CC, such as in astronomy and homeland security, demand high spatial resolution in the reconstructed images. Spatial resolution is typically determined by the uncertainty in the positions of Compton scattering events within the camera, which depends on the size of the camera’s voxels. Enhancing resolution typically requires a large number of small voxels, which in turn requires a high number of readout channels. Classical multiplexing measurement methods are commonly used to increase the number of measured voxels. However, these methods require passive demodulation networks to identify the voxel within each group where the event occurred. Such methods introduce performance degradation, occupy additional space on the printed circuit board, and increase system complexity. In this study, we introduce an algorithmic framework and a numerical proof-of-concept of a novel approach to enhance spatial resolution by dividing the volume of a CC into additional voxels without increasing the number of readout channels. To achieve this, we design the CC such that each readout channel is connected to a group of voxels, without the need for passive demodulation networks. The proposed voxels grouping introduces an inherent ambiguity in the acquired events, as it becomes impossible to determine the specific voxel within each group in which an event occurred. To overcome this limitation, we introduce an adaptation of conventional 2D image reconstruction, based on the widely-used Maximum Likelihood Expectation Maximization (MLEM), extending it to a grouping-aware framework, yielding a <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Grouped MLEM</i> algorithm. We extensively evaluate the resolution enhancement provided by our voxels grouping method through simulations, considering setups with single and two adjacent <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">137</sup>Cs gamma sources. The results systematically demonstrate that the proposed voxels grouping method, combined with the Grouped MLEM algorithm, improves gamma-ray source localization and separation without requiring additional readout channels. This novel approach holds the potential to enable the development of high spatial resolution, cost-effective, and mobile CCs.