Experimental verification of a technique for predicting scattered radiation transfer: application to low photon...

It is shown that predictions can be made of scatter-to-primary ratios for a variety of mammographic configurations. The different configurations can be produced by changes in source-detector distance; source-phantom distance; air gap; photon energy; phantom composition, thickness, and cross-sectional area. A detailed analysis of the effect of a grid is also considered. The predictions are not computationally intensive. Experimental verifications of the predictions for varying phantom cross-sectional dimensions, source-detector distance, and air gaps with and without a grid have been carried out. Variations with phantom thickness were not considered in this paper. Detailed comparisons between experiment and theory indicate that scatter-to-primary ratios were predicted within one standard deviation and the coefficient of variation is within 3% for most of the data points with the worst case coefficient of variation about 6%. Measurements of grid secondary transmission have also been compared to theoretical predictions and agreement is within the coefficient of variation of about +/- 15%.