BOMAB phantom variability: do small dimensional changes matter?

The Human Monitoring Laboratory (HML) has had a number of BOMAB phantoms built over the years. Upon characterization, it has been found that the dimensions of the phantoms are always slightly different. This study has looked at the effect of these small variances in dimensions of the phantoms and compared the results to what is required in the industry standard using Monte Carlo simulations for three counting geometries: the HML's scanning detector whole body counter, the StandFast whole body counter, and the W-chair whole body counter. It has been found that the effect of these small variations on the performance of these phantoms is very minor (<5%). It is reassuring to find that small variations in manufacturing, even if individual sections are non-compliant, have such a minor effect on performance as to be considered a negligible effect for any counting system's geometry.

Can the HML's sliced BOMAB phantom be used in any whole body counter with a reduced number of sources?

The sliced Bottle Manikin Absorber (BOMAB) phantom was originally proposed as an alternative to a commercially available phantom, but it suffers from the disadvantage of containing over 160 sources that need to be manufactured; however, it was found that the number of slices could be reduced substantially and that two slices in the sliced phantom gave the same performance characteristics over a wide energy range as a conventional BOMAB phantom for a particular counting system. This work explores the adaptability of this phantom to another counting geometry. The response of the Human Monitoring Laboratory's whole-body counter measuring this phantom with a decreasing number of planar sources has been modelled using MCNP5 over a wide energy range (122-2754 keV). It was found that the best agreement was obtained when the phantom contained 10 sources, 1 in the mid point of each section. As this is a different result from a previous finding, any other counting geometry will have to be assessed to determine the optimum loading if the sliced phantom is to be used. Also, it is clear that this type of phantom cannot be used for an intercomparison that will encounter different counting geometries, unless it contains a full loading of sources.