Abstract
P377
Introduction: Targeted alpha therapy (TAT) has been proposed for cancer treatment in the past decades and has gained growing interest in clinical studies after the success of the FDA-approved radionuclide: 223Ra dichloride. More efficient killing of the malignant cells and less damage to normal tissue make alpha emitters better candidates than beta emitters in targeted radionuclide therapy. Several radionuclides such as Actinium-225 (225Ac), Bismuth-213 (213Bi), Astitine-211 (211At), and Lead-212 (212Pb) combined with antibodies or small molecules are under investigation in clinical trials, which leads to the requirement for multi-energy gamma imaging for quality control over the course of treatment.
We are developing a Compton scattering cone-based gamma imaging concept using electronic collimation, which enables multi-energy gamma photon imaging with high efficiency compared with mechanically collimated SPECT. The concept of the Compton camera (CC) for clinical applications was first proposed in the 1970s and multi-gamma imaging capability has since gained more attention. This work focuses on evaluating the capability of multi-energy gamma imaging with an optimized CC system, using GATE Monte Carlo simulation toolkit.
Methods: The simulated Compton camera is composed of one bismuth germanate (BGO) absorber (72x72x2 cm3) and multi-layer scatterer (48x48x0.2 cm3). Two materials of the scattering layers were compared: silicon and cadmium zinc telluride (CZT). We compared the imaging resolution of different geometry design at 440 keV with varying detector spatial resolution, varying number of scattering layers and varying distances between scatterer and absorber. A simulated Derenzo phantom in air having spheres with diameter 2 mm, 1.6 mm, 1.2 mm, 1.1 mm, 0.9 mm and 0.7 mm, was placed 5 cm from the scatterer. The central cut of the reconstruction and detailed system geometry are shown in the supporting data. The source containing 1.6 x105 Bq in total was simulated in air for 100 seconds. We employed the list mode MLEM reconstruction algorithm to evaluate the imaging resolution. To show the 3D resolution of the system, an additional test was added in which the source was placed parallel to the frontal plane. After geometry optimization, we chose a single CZT scattering layer with 0.5 mm detector resolution and BGO absorber with 1.25 mm resolution to evaluate the gamma imaging of 225Ac, with the source placed at 20 cm from the camera. Two energy peaks (218 keV, 440 keV) were simulated separately.
First for design optimization, we reconstructed images of 213Bi (440 keV) with different system geometries. Once arrived at optimized geometry, we reconstructed images of 225Ac for two energy peaks (218 keV, 440 keV) separately to show prospective imaging performance.
Results: We show that with the optimized geometry, <2.0 mm resolution can be obtained at 218 keV and 440keV with limited counts (source placed 20 cm from the camera), and 1.1 mm resolution can be obtained at 440 keV (source placed 5 cm from the camera). Compared with one single layer scatterer, detection sensitivity with multi scattering layers is larger, but the reconstructed image quality is not improved. A 10 cm distance between the scattering layers and the absorber leads to a better resolution than a 5 cm separation.
Conclusions: The proposed CC system can provide < 2.0 mm resolution at energies 218 keV and 440 keV for whole body imaging, and 1.1 mm resolution at 440 keV for small animal imaging. Further evaluation of the imaging performance for human whole body is ongoing, and a simulation and reconstruction with more realistic data will be presented in the conference. Currently, no energy resolution was modeled for image reconstruction, but we expect with energy blurring the image resolution degrades especially for whole body human imaging which will be investigated thoroughly and will be presented during the conference.
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.