Establishing a quality control program for an automated dosimetry system

The assessment of fetal dose from a special procedure in the clinical environment is difficult. as patient size, fluoroscopic beam motion, and imaging sequences vary significantly from study to study. Fetal dose is particularly difficult to estimate when the fetus is exposed partially or totally to scatter radiation from images taken in other locations of the motherʹs body. A method to reliably estimate fetal dose has been developed by using template based input files for the Monte Carlo radiation transport code MCNP. Female patient phantoms at 0, 3, 6, and 9 months of pregnancy and source terms for common diagnostic tube potentials are used to rapidly build an input file for MCNP. The phantoms can be easily modified to fit patient shape. The geometry and beam location for each type of image acquired (i.e. fluoroscopy, spot filming, etc.) is verified by the use of a 3D visualization code (Sabrina). MCNP is then run to estimate the dose to the embryo/fetus and the exposure to skin entrance (ESE) for the beam being modeled. The actual ESE for the beam is then measured with ion chambers and the fetal dose is determined from the MCNP supplied ratio of ESE to fetal dose. Runs are made for each type of imaging and the doses are summed for the total fetal dose. For most procedures, the method can provide an estimate of the fetal dose within one day of the study. The method can also be used to prospectively model a study in order to choose imaging sequences that will minimize fetal dose. © 1999 American Association of Physicish in Medicine. [S0094-2405(99)02208-7]