Purpose: Skeletal survey is an important tool in evaluating suspected physical abuse in preverbal children (0-3 years). The need to demonstrate subtle injuries demands high contrast and spatial resolution. This is especially true for the imaging of infant skulls. The goal of this study was to develop radiographic techniques based on head thickness and the target exposure indices (EI_T) needed for clinical diagnosis.
Methods: The study included skull radiographs of 30 patients (aged 0-32 months) who underwent pediatric skeletal surveys using one type of digital system. All images were taken table-top using manual techniques without grid. From these images, head thicknesses were measured. Exposure techniques and EI were also extracted. Most images were taken at 60 kV, with the rest at 65 kV. For the latter group, an equivalent mAs needed at 60 kV was calculated following the 15% rule. Because EI scales linearly with mAs, a new mAs was then calculated for each image to provide a hypothetical EI_T of 800 (mAs_800). Regression analysis was performed to relate mAs_800 to head thickness. From this relationship, the median head thickness of each age group was used to calculate mAs_800, which was then scaled to obtain the mAs needed for a clinically desired EI_T.
Results: The natural logarithm of mAs_800 increased linearly with head thickness (R²=0.38-0.59). For the same head thickness, a higher mAs was needed in the lateral direction. Evaluation of clinical image quality by experienced pediatric radiologists showed that EI_T of 2000 was desirable for newborns who have small and underdeveloped skulls, whereas EI_T of 1000 was adequate for toddlers. The relationship between mAs_800 and head thickness allowed a new technique chart to be developed to provide such age-dependent EI_T.
Conclusion: The regression relationship between mAs_EI_T and anatomical thickness may be exploited to develop manual technique charts for children.
Radiography, Noise, Optimization