Monte Carlo study of a double scattering Comptoncamera with GEANT4

Abstract

The Compton camera is a very promising imaging device in nuclear medicine considering its compactness, significantly low cost, low patient dose, and inherently simple image reconstruction process. The current version of Compton cameras uses two position-sensitive detectors, i.e., scatterer and absorber. The scatterer determines the location of the Compton scattering and the deposited energy in the scatterer while the absorber determines the location and energy of the absorbed photon in the absorber. The size of the absorber segment, especially the thickness, is kept relatively large to provide an adequate level of detection efficiency, which on the other hand seriously limits the resolution of the Compton camera. In this study, we wanted to improve the resolution of the Compton camera by utilizing two scatterers, instead of one, and one absorber. The new system is composed of two double-sided silicon strip detectors (5 cm x 5 cm x 0.15 cm, 16 strips on each side) as scatterers and one coaxial-type high-purity germanium detector (7 cm diameter, 7 cm length) as absorber. To test our idea, the double scattering Compton camera was modeled using GEANT4 simulation toolkit. Our simulation model includes the details of the detector geometry, energy resolution, spatial resolution (i.e., detector segmentation), energy discrimination level of the scatterers and absorber. Doppler energy broadening was included by implementing the Penelope physics model in the GEANT4 toolkit. Our Monte Carlo simulation result is encouraging; the resolution was improved almost by factor of two by using two scatterers instead of one. The resolution of the double scattering Compton camera is currently estimated 10.4 mm (= FWHM) for 18F point source at 6 cm from the camera and it will be further improved through the optimization of the system. It was found that the segmentation of the detectors (especially the front scatterer) still most significantly affects the resolution of the double scattering Compton camera (FWHM = 8.75 mm). Note that the Doppler energy broadening and energy resolution of the detectors result in FWHM of only 3.04 mm and 3.15 mm, respectively. This result suggests that the use of a more segmented detector for the front scatterer will effectively improve the resolution of the Compton camera. A prototype system is now under construction at Hanyang University in Korea to demonstrate the capabilities of the double scattering Compton camera.