Grid removal and impact on population dose in full-field digital mammography.

G. Gennaro,Luc Katz,H. Souchay,R. Klausz,Claudio Alberelli,C. di Maggio
DOI: https://doi.org/10.1118/1.2426402
2007-02-01
Abstract:The study purpose was to determine the impact of anti-scatter grid removal on patient dose, in full field digital mammography. Dose saving, phantom based, was evaluated with the constraint that images acquired with and without grid would provide the same contrast-to-noise ratio (CNR). The digital equipment employed a flat panel detector with cesium iodide for x-ray to light conversion, 100 microm pixel size; the x-ray source was a dual-track tube with selectable filtration. Poly(methyl-emathocrylate) (PMMA) layers in the range 20-70 mm were used to simulate the absorption of different breast thickness, while two Al foils, 0.1 and 0.2 mm thick were used to provide a certain CNR. Images with grid were acquired with the same beam quality as selected in full automatic exposure mode and the mAs levels as close as possible, and the CNR measured for each thickness between 20 and 70 mm. Phantom images without grid were acquired in manual exposure mode, by selecting the same anode/filter combination and kVp as the image with grid at the same thickness, but varying mAs from 10 to 200. For each thickness, an image without aluminum was acquired for each mAs value, in order to obtain a flat image to be used to subtract the scatter nonuniformity from the phantom images. After scatter subtraction, the CNR was measured on images without grid. The mAs value that should be set to acquire a phantom image without grid so that it has the same CNR as the corresponding grid image was calculated. Therefore, mAs reduction percentage was determined versus phantom thickness. Results showed that dose saving was lower than 30% for PMMA equivalent breast thinner than 40 mm, decreased below 10% for intermediate thickness (45-50 mm), but there was no dose gain for thickness beyond 60 mm. By applying the mAs reduction factors to a clinical population derived from a data base of 4622 breasts, dose benefit was quantified in terms of population dose. On the average, the overall dose reduction was about 8%. It was considered small, not sufficient to justify a clinical implementation, and the anti-scatter grid was maintained.
What problem does this paper attempt to address?