Program Information
Evaluation of Intra-Organ Dose Heterogeneity Using XCAT Phantoms
J Fenoli*, J Hoye, S Sharma, J Spencer, B Harrawood, P Segars, E Samei, A Kapadia, Duke University Medical Center, Durham, NC
Presentations
SU-H2-GePD-I-5 (Sunday, July 30, 2017) 3:30 PM - 4:00 PM Room: Imaging ePoster Lounge
Purpose: Dose values for radiography and CT are typically reported as effective, whole-body doses, with a subset of those studies reporting total organ doses. Even fewer studies have investigated the variation of dose within organs, as this is nearly impossible to measure and difficult to simulate in phantoms. Using realistic human (XCAT) phantoms and Monte-Carlo simulations, it is possible to perform this level of analysis. Previous simulations have used similar methods to optimize CT routines and calculate overall organ dose, but no studies have investigated the distribution of intra-organ dose. The purpose of this work is to evaluate intra-organ dose heterogeneity using realistic XCAT phantoms.
Methods: A Geant4 simulation was used to model a uniformly-distributed, 100 keV photon beam incident on the chest of a male XCAT phantom. The beam area included the heart, lungs, kidney, liver, gall bladder, spleen, stomach, pancreas, adrenal glands, thymus, trachea/bronchi, and esophagus. The phantom was modeled with uniform voxel size of 3.45 mm along each dimension. Average organ dose was calculated for each organ in the phantom. Dose-volume histograms (DVH) were created for 12 organs of interest in the abdominal region and compared against the average organ dose values.
Results: The DVH curves showed heterogeneity in all organ doses, with some organs receiving up to 6 times the average dose. Lung showed maximum heterogeneity due to its low relative density. Excluding the lung, the standard deviation of the DVH between organs did not exceed 6.3% for any dose level.
Conclusion: The results demonstrate the ability to evaluate organ-dose heterogeneity for clinical diagnostic imaging using a Monte-Carlo approach.
Contact Email: