Cynthia H. McCollough PhD, Presenter: Research grant, Siemens AG

Jodie A. Christner PhD, Abstract Co-Author: Grant, Siemens AG

Vanessa Rueda, Abstract Co-Author: Nothing to Disclose

Juan Carlos Ramirez Giraldo MS, Abstract Co-Author: Nothing to Disclose

Thomas J. Vrieze RT(R), Abstract Co-Author: Nothing to Disclose

Shuai Leng PhD, Abstract Co-Author: Nothing to Disclose

To predict the radiation dose in the center of the scan volume for any patient size using scanner output information (CTDIvol).

The effect of size on patient dose was studied using 8 tissue-equivalent abdominal phantoms having lateral widths from 10 – 40 cm, representing newborn to large patients. Additional attenuating material was placed adjacent to the abdominal phantoms to simulate a complete torso, where torso length was proportionally adjusted to lateral width. Abdominal spiral, sequential and perfusion scans were performed using clinical scan parameters at both 80 and 120 kV. Dose to the center and four surface locations were measured at the center of the scan range according to AAPM Report 111 using a small ion chamber (0.6 cm3, Radcal). Measured doses were normalized to scanner output for the same technique factors, using the CTDIvol from the scanner console. The normalized center, mean surface, and mean dose (1/3 x center dose + 2/3 x mean surface dose) were fit to several measures of phantom size (anterior-posterior (AP), lateral, or AP + lateral dimensions; cross-sectional area; perimeter) using an exponential relationship. Regression analysis was used to determine the ability of an exponential relationship to predict size-specific doses from scanner output and patient size.

After normalization to CTDIvol, all dose values (center, mean surface or their weighted average) demonstrated excellent correlation to an exponential relationship for each measure of patient size (R2 > 0.98). 80 and 120 kV regression coefficients were similar for center doses, but varied markedly for surface dose, and hence mean dose. Surface dose from perfusion acquisitions was nearly a factor of two lower than from spiral or sequential abdominal scans after normalization to CTDIvol, reflecting CTDIvol’s known overestimation of skin dose for perfusion exams.

Center, surface and mean dose at the center of the scan range, after normalization to CTDIvol, were exponentially related to patient size. Regression equations were strongly correlated to data measured over 8 patient sizes, ranging from newborn to a large adult. Thus, a standardized, absolute measure of scanner output (CTDIvol) can reliably estimate the center, surface or mean dose for any patient size.

Currently-reported measures of scanner output (CTDIvol) can reliably estimate dose to the patient for any patient size.

McCollough, C, Christner, J, Rueda, V, Ramirez Giraldo, J, Vrieze, T, Leng, S, Estimating Patient-specific Dose from Scanner Output (CTDIvol): Yes We Can!.  Radiological Society of North America 2010 Scientific Assembly and Annual Meeting, November 28 - December 3, 2010 ,Chicago IL. http://archive.rsna.org/2010/9007169.html