Evaluation of two scatter corrections techniques for an uncollimated transmission flood source

Abstract

Scattered photons can be a problem in transmission computed tomography (TCT) when employing an uncollimated transmission source. As an uncollimated transmission source can generate many scatter events in the transmission data, accurate scatter correction is necessary during transmission imaging. The aim of this study was to evaluate two scatter correction techniques which can be used with an uncollimated flood source for transmission imaging. Energy spectra of 140keV-photons transmitted through different thicknesses of attenuating material were acquired at different source detector distances (SDD) for each pixel using a scintillation camera. The scatter contribution to each energy spectrum was determined by subtracting the primary counts from the total counts for a specific thickness of material. The scatter percentage (SP) in the 20% energy window was calculated as the ratio of scattered to total counts and expressed as a percentage. These values were compared with the SP values that were calculated with the triple energy window (TEW) and photon energy recovery (PER) scatter correction techniques. The SP decreases as the SDD increases. The SP obtained with the TEW and PER scatter correction techniques for the varying thicknesses of Lucite compare favorably with those calculated for the open transmission source at distances of 20, 40, 60 and 80cm from the detector. It is evident from the results that both scatter correction techniques can be used successfully during transmission imaging with an uncollimated flood source. The TEW scatter correction tends to overestimate the low scatter values and underestimate high scatter values. The TEW scatter correction technique is easier to implement than the PER technique during clinical imaging. As a result of the accuracy of the PER scatter correction technique it is the method of choice for scatter correction for transmission imaging. © 2009 Springer-Verlag.