The effect of added filtration on radiation dose and image quality in digital radiography of newborns.

Abstract

Repeated chest and/or abdomen radiographs are needed on mobile digital radiography (DR) units to assess infants' health status. Optimization of DR tube peak kilovoltage (kVp) and tube current time product (mAs) to derive images of diagnostic quality at as low as reasonably achievable radiation dose has been a challenging task. To investigate the effect of exposure parameters and additional filtration on entrance skin dose and image quality in newborn DR imaging. A physical anthropomorphic phantom that simulates the average full-term neonate was used. Chest and chest/abdomen DR images were acquired at the manufacturer's recommended kVp/mAs exposure parameters followed by a series of image acquisitions at different kVp/mAs and beam filtration combinations. Entrance skin dose (ESD) and signal difference to noise ratio (SdNR) for soft tissue, bone and feeding gastric tube were estimated in raw unprocessed images. A figure of merit (FOM) analysis provided the kVp/mAs and filtration that generated images of adequate quality at the lowest ESD. Signal difference increased with kVp and progressively decreased with increasing filtration. Compared to the manufacturer's recommended 53 kVp/1.6 mAs exposure parameters, ESD was reduced by 76% (47.61μGy versus 11.3μGy) in the chest and 66% (47.61μGy versus 16.14μGy) in the chest/abdomen when the exposure parameters and additional beam filtration provided by the FOM analysis were used. The results derived from this phantom study suggest that additional beam filtration along with the appropriate leverage of exposure parameters can lower ESD while maintaining image quality in full-term newborns.