SU‐FF‐I‐28: A Critical Look at the Numerical Coefficients in CTDIVOL

Purpose: To critically examine the assumptions leading to the formula for CTDI VOL , which uses measured CTDI values at the center and periphery to estimate the overall average dose to a homogeneous cylindrical phantom. Method and Materials: CTDI VOL is a widely used figure of merit that estimates the average dose delivered by a CT scan to standard homogeneous cylindrical cylinders and offers a means of comparing doses delivered by different machines, or by the same machine using different scanning parameters. The calculation of CTDI VOL requires the measurement of two CTDI values: CTDI E measured at the edge of the cylinder and CTDI C measured at the center. CTDI VOL= ( a ⋅ CTDI E+ b ⋅ CTDI C) / Pitch . The following three assumptions are made in order to determine a and b : 1): The phantom is a homogeneous cylinder positioned coaxially with the gantry. 2) The cylinder is exposed to a constant beam of radiation during the rotation of the gantry. 3) At any one instant during the scanner's rotation, the radiation density within the cylinder varies smoothly with position. Following a 360° rotation, assumptions 1 – 3 imply that the dose distribution is also cylindrically symmetric, smoothly varying, and at an extremum at the center. Results: The simplest result following from assumptions 1 – 3 is that a = b = 1/2. These do not agree with the values of a and b (2/3 and 1/3, respectively) that are used conventionally; these conventional values have the unphysical implication that the dose gradient at the center is discontinuous. Conclusion: CTDI VOL , as it is currently calculated, underestimates the relative contribution made by the central CTDI value to the average dose, leading to a significant systematic error in a number used by regulatory and accreditation agencies. Equal weighting provides a more accurate value and should be employed.