Sung-Woo Lee PhD, Presenter: Nothing to Disclose

Sunyoung Jang PhD, Abstract Co-Author: Nothing to Disclose

Bruce H. Curran MEng, Abstract Co-Author: Nothing to Disclose

Jong Oh Kim PhD, Abstract Co-Author: Nothing to Disclose

Edward Sternick PhD, Abstract Co-Author: Nothing to Disclose

David E. Wazer MD, Abstract Co-Author: Nothing to Disclose

Purpose/Objective(s): CT Scanning is now the routine procedure for acquiring patient data in radiation therapy. When tissue heterogeneity is taken into account, most treatment planning systems (TPS) store the relationship between the acquired CT number and electron density in a table associated with the particular CT system used for image acquisition. Different CT acquisition techniques, however, can alter the CT number - electron density relationship and thus change the dose calculation to both target and organ-at-risk (OAR) structures. In this study, the effects of changes in CT number as a function of voltage, power, and acquisition technique (helical vs axial) on dosimetric planning are investigated Materials/Methods: A Brilliance Big Bore CT system with 16-slice capability (Philips Medical Systems, Andover, MA) was used for image acquisition. An Eclipse TPS (Varian Medical System, Palo Alto, CA) with a fast Monte Carlo algorithm (XVMC) and a Pinnacle TPS (Philips Medical Systems, Andover, MA) with Collapsed Cone Convolution (CCC) algorithm were used to study the variation in calculated dose. An electron density phantom (Gammex 467, Gammex-RMI, Middleton, WI) was used for determination of the CT number vs electron curves. Six different CT number vs electron density curves were generated using different combinations of scanning technique (helical, axial), voltage (90 kVp, 120 kVp) and power (100 mAs, 300 mAs). Each of these curves was entered into both TPSs and then used in calculation of an AP/PA pelvis treatment plan. Treatment beams using 6 MV, 10 MV, and 18 MV were used for the dose calculations to 4 different target volumes and two OARs. Dose Volume Histograms were generated and used in the evaluation of the results. Results: No significant differences in CT number were observed between helical and axial scanning techniques for tissues up to the density of cortical bone (?=1.69 g/cm3). The maximum dose difference observed was less than 4.4% and no changes were found in the calculated MU settings under conditions of similar voltage and mAs. Differences were found when a 90 kVp acquisition technique was used instead of the more common 120 kVp. In these cases, calculated MUs were different by -4.5%, -3.8%, and -3.1% for 6 MV, 10 MV, and 18 MV beams respectively. Conclusions: A widely accepted metric for dose calculations is achieving an accuracy of 5% or better. In this study, calculated doses and MUs were all within this limit, indicating that, for normal tissues, CT acquisition techniques will not significantly alter clinical results. However, this study encompassed conventional planning/delivery techniques and did not include variations that may be induced with higher density materials such as might be found in hip prostheses or dental implants.

Lee, S, Jang, S, Curran, B, Kim, J, Sternick, E, Wazer, D, Do Variations in CT Scanning Techniques Affect Dosimetric Calculations with Advanced Dose Calculation Algorithms?.  Radiological Society of North America 2009 Scientific Assembly and Annual Meeting, November 29 - December 4, 2009 ,Chicago IL. http://archive.rsna.org/2009/8501782.html