Monte Carlo simulations in CT for the study of the surface air kerma and energy imparted to phantoms of varying size and position

A Monte Carlo computational model of CT has been developed and used to investigate the effect of various physical factors on the surface air kerma length product, the peak surface air kerma, the air kerma length product within a phantom and the energy imparted. The factors investigated were the bow-tie filter and the size, shape and position of a phantom which simulates the patient. The calculations show that the surface air kerma length product and the maximum surface air kerma are mainly dependent on phantom position and decrease along the vertical axis of the CT plane as the phantom surface moves away from the isocentre along this axis. As a result, measurements using standard body dosimetry phantoms may underestimate the skin dose for real patients. This result is specially important for CT fluoroscopic procedures: for an adult patient the peak skin dose can be 37% higher than that estimated with a standard measurement on the body AAPM (American Association of Physicists in Medicine) phantom. The results also show that the energy imparted to a phantom is mainly influenced by phantom size and is nearly independent of phantom position (within 3%) and shape (up to 5% variation). However, variations of up to 30% were found for the air kerma to regions within the AAPM body phantom when it is moved vertically. This highlights the importance of calculating doses to organs taking into account their size and position within the gantry.