The UF family of pediatric tomographic models

2005 
The use of computed tomography (CT) in pediatric radiology has rapidly increased over the past 15 years. While CT accounts for only about 10% of diagnostic radiological exams, it contributes disproportionately to the patient collective dose. The issue of radiation risk is of far greater importance in pediatric radiology as children are inherently more radiosensitive than adults. Strategies for dose reduction, as well as optimization of image quality versus patient dose, are thus crucial in pediatric CT. While many indicator quantities are easily measured in the clinic (such as the CTDI or DLP), the effective dose is the most fundamental indicator of patient risk in pediatric CT, a quantity requiring knowledge of individual organ doses. The effective dose thus requires the use of heterogeneous anthropomorphic models of pediatric anatomy. While many tomographic models of adult anatomy have been developed over the past 15 years, relatively few models are of children as needed in medical CT dosimetry. Using both cadaver and live patient image sets, we have been developing a series of tomographic pediatric models for use in pediatric radiology. The current UF family of pediatric models currently stands at six models: 6-day female, 9-month male, 4-year female, 8-year female, 11-year male, and 14-year male. The newborn model was constructed from a cadaver scan, while the remaining models were developed from segmentation of head and chest-abdomen-pelvis (CAP) scans of live patients. In addition to their use in Monte Carlo simulations of CT, fluoroscopy, and projection radiography, physical phantoms of the UF newborn and UF 9-mo models have been constructed using tissue equivalent materials for soft tissue, lung, and bone. Internal placement of high-sensitivity MOSFET dosimeters at organ centroids, and Monte Carlo-derived point-to-organ dose scaling factors provide a means of physical verification of doses estimates via computer simulation.
    • Correction
    • Cite
    • Save
    • Machine Reading By IdeaReader
    12
    References
    0
    Citations
    NaN
    KQI
    []