Use of beam stoppers to correct random and scatter coincidence in PET: A Monte Carlo simulation

Abstract 3D acquisition of positron emission tomography (PET) produce data with improved signal-to-noise ratios compared with conventional 2D PET. However, the sensitivity increase is accompanied by an increase in the number of scattered photons and random coincidences detected. Scatter and random coincidence lead to a loss in image contrast and degrade the accuracy of quantitative analysis. This work examines the feasibility of using beam stoppers (BS) for correcting scatter and random coincidence simultaneously. The origins of the photons are not on the path of non-true event. Therefore, a BS placed on the line of response (LOR) that passes through the source position absorbs a particular fraction of the true events but has little effect on the scatter and random events. The subtraction of the two scanned data, with and without BS, can be employed to estimate the non-true events at the LOR. Monte Carlo (MC) simulations of 3D PET on an EEC phantom and a Zubal Phantom are conducted to validate the proposed approach. Both scattered and random coincidences can be estimated and corrected using the proposed method. The mean squared errors measured on the random+scatter sinogram of the phantom obtained by the proposed method are much less than those obtained using the conventional correction method (the delayed coincidence subtraction for random correction combined with single scatter simulation for scatter correction). Preliminary results indicate that the proposed method is feasible for clinical application.