Scintillation Event Localization in Hemi-Ellipsoid Detector for SPECT, a simulation study using Geant4 Monte-Carlo

TL;DR

This simulation study demonstrates that a hemi-ellipsoid detector in SPECT can accurately localize scintillation events with sub-millimeter precision using Geant4, enabling high-resolution cardiac imaging.

Contribution

The paper introduces a novel Monte Carlo simulation approach for scintillation event localization in a hemi-ellipsoid detector, achieving high accuracy in a SPECT system.

Findings

Localization error averaged 0.71 mm across zones

Back-projection errors were less than 1 mm

High resolution achieved with accurate event localization

Abstract

A high sensitivity Cardiac SPECT system using curved crystals with pinhole collimation was proposed previously (Dey, IEEE TNS 2012, Bhusal et al, Med. Phys. 2019). Here, we hypothesize that a high curvature hemi-ellipsoid detector results in measurable differences in light distribution from events at different depths in the crystal. This was tested by analyzing the scintillation light in hemi-ellipsoid detector using Monte-Carlo (Geant4) and evaluation of both the localization error at detector and the back-projected errors in object-space. We used Geant4 to simulate the propagation of scintillation light in a monolithic hemi-ellipsoidal CsI crystal. A look-up table (LUT) was created to map the points inside the crystal to the expected light pattern on the crystal surface using Geant4. In thirteen zones across the crystal, gamma-rays were simulated and the resulting scintillator light intensity on the surface was captured, serving as our experimental interactions. A Poisson-statistics-based algorithm was developed to limit the search of the gamma-ray event locations into small regions of the LUT and fine-tuned by interpolating between selected LUT points by comparing the light distribution of the gamma interactions and LUT light patterns. The localized events were individually back-projected to the object mid-plane, and the errors recorded. Excluding some outliers (up to 2%), the localized errors averaged over all the zones was 0.71 (+/-0.44) mm with a worse case of 1.36 (+/-0.67) mm at the apex. When back-projected to the midplane of the object for Cardiac SPECT, the errors were <1mm and average error was 0.4(+/-0.22) mm, due to the high system magnification afforded by the detector. Thus, for our high sensitivity system, we were also able to achieve high resolution, assuming perfect pinhole collimator resolution recovery for Cardiac SPECT application.