Measurement of scattered radiation in a volumetric 64-slice CT scanner using three experimental techniques

Compton scatter poses a significant threat to volumetric x-ray computed tomography, bringing cupping and streak artefacts thus impacting qualitative and quantitative imaging procedures. To perform appropriate scatter compensation, it is necessary to estimate the magnitude and spatial distribution of x-ray scatter. The aim of this study is to compare three experimental methods for measurement of the scattered radiation profile in a 64-slice CT scanner. The explored techniques involve the use of collimator shadow, a single blocker (a lead bar that suppresses the primary radiation) and an array blocker. The latter was recently proposed and validated by our group. The collimator shadow technique was used as reference for comparison since it established itself as the most accurate experimental procedure available today. The mean relative error of measurements in all tube voltages was 3.9 +/- 5.5% (with a maximum value of 20%) for the single blocker method whereas it was 1.4 +/- 1.1% (with a maximum value of 5%) for the proposed blocker array method. The calculated scatter-to-primary ratio (SPR) using the blocker array method for the tube voltages of 140 kVp and 80 kVp was 0.148 and 1.034, respectively. For a larger polypropylene phantom, the maximum SPR achieved was 0.803 and 6.458 at 140 kVp and 80 kVp, respectively. Although the three compared methods present a reasonable accuracy for calculation of the scattered profile in the region corresponding to the object, the collimator shadow method is by far the most accurate empirical technique. Nevertheless, the blocker array method is relatively straightforward for scatter estimation providing minor additional radiation exposure to the patient.