Gamma-ray imaging facilitates the efficient detection, characterization, and localization of compact radioactive sources in cluttered environments. Fieldable detector systems employing active planar coded apertures have demonstrated broad energy sensitivity via both coded aperture and Compton imaging modalities. However, planar configurations suffer from a limited field of view, especially in the coded aperture mode. To improve upon this limitation, we introduce a novel design by rearranging the detectors into an active coded spherical configuration, resulting in a 4π isotropic field of view for both coded aperture and Compton imaging. This paper focuses on the low-energy coded aperture modality and the optimization techniques used to determine the optimal number and configuration of 1-cm3 CdZnTe coplanar grid detectors on a 14-cm diameter sphere with 192 available detector locations.

Background

Targeted alpha-particle therapy (TAT) has great promise as a cancer treatment. Arguably the most promising TAT radionuclide that has been proposed is ²²⁵ Ac. The development of ²²⁵ Ac-based radiopharmaceuticals has been hampered due to the lack of effective means to study the daughter redistribution of these agents in small animals at the preclinical stage.

Purpose

The ability to directly image the daughters, namely ²²¹ Fr and ²¹³ Bi, via their gamma-ray emissions would be a boon for preclinical studies. That said, conventional medical imaging modalities, including single photon emission computed tomography (SPECT) based on nonmultiplexed collimation, cannot be employed due to sensitivity limitations.

Methods

As an alternative, we propose the use of both coded aperture and Compton imaging with the former modality suited to the 218-keV gamma-ray emission of ²²¹ Fr and the latter suited to the 440-keV gamma-ray emission of ²¹³ Bi.

Results

This work includes coded aperture images of ²²¹ Fr and Compton images of ²¹³ Bi in tumor-bearing mice injected with ²²⁵ Ac-based radiopharmaceuticals.

Conclusions

These results are the first demonstration of visualizing and quantifying the ²²⁵ Ac daughters in small animals through the application of coded aperture and Compton imaging.

The ability to localize and map the distribution of gamma-ray emitting radionuclides in 3D has applications ranging from medical imaging to nuclear security. In the case of radiological source search and nuclear contamination remediation, the deployment of freely moving detection systems such as handheld instruments or ground/aerial-based vehicles is critical in overcoming the inverse square law and complex shielding scenarios. Using contextual sensors, these systems can simultaneously generate 3D maps of the surrounding environment and track the position and orientation of the gamma-ray sensitive detectors in that environment. The fusion of contextual scene data and gamma-ray detector data to facilitate real-time 3D gamma-ray image reconstruction has been demonstrated with mobile high purity germanium (HPGe) and CdZnTe-based Compton cameras for gamma-ray energies ranging from a few hundred keV to several MeV. Here we apply this approach for lower energy (50-400 keV) gamma rays, using a handheld CdZnTe-based omnidirectional imaging system and an active coded mask imaging modality. We present our approach to real-time reconstruction using a scene-data-constrained graphics processing unit (GPU)-accelerated list-mode maximum likelihood expectation maximization (MLEM) algorithm and show results from several measurements in the lab and in the field.

The analysis and interpretation of coincidence events in a Compton camera requires the comparison of the expected rates of observed events from sources with various emission rates, energy spectra and spatial distributions. Radioactive source distributions are often represented by the activity distributed among numerous voxels; each voxel having uniform internal activity and spectra within a cube. In this paper a mathematical model is constructed that predicts the expected rate of coincident Compton events from the rate of emissions from a single voxel source. This detailed model incorporates (1) the finite voxel size, (2) the blurring of the “Compton cone” by the limitations of energy resolution in the detectors and (3) the uncertainty in the Compton cone-axis due to the limited spatial resolution and ‘lever-arm’ separation between the coincident interactions. The resultant rates can be used to generate the system response matrix for source reconstruction and, therefore, are directly applicable in list-mode MLEM source reconstruction algorithms.