Diameter measurement of vascular model on CT angiography using model-based iterative reconstruction: effect of tube current on accuracy.

Abstract

The purpose of this study was to evaluate the effect of tube current on the accuracy of vascular diameter measurements on CT angiography using model-based iterative reconstruction (MBIR). We constructed a physical phantom composed of nine vascular models of three wall thicknesses filled with contrast material of three densities and scanned the phantom using a 64-MDCT unit with tube currents of 80, 40, 20, and 10 mA. We reconstructed raw data using MBIR and filtered back projection (FBP) algorithms and examined image sharpness by edge rise distance using four tube currents in each of the two reconstruction methods. We measured the mean inner diameter for each model for each set of image data using MBIR with the four tube currents (80, 40, 20, and 10 mA) and FBP with tube current of 80 mA. Sharpness decreased as tube current decreased using MBIR (p < 0.0001) but did not differ significantly among the four tube currents using FBP (p = 0.0506). Sharpness using MBIR with 10 mA tube current was comparable to or significantly worse than that with FBP. The measurement error using MBIR tended to increase as tube current decreased. The measurement errors using MBIR with 10 mA tube current were comparable to or significantly larger than those using FBP for eight of the nine vascular models. By decreasing spatial resolution, MBIR with lower tube current can enhance errors in measurement of vascular diameter and should be applied carefully in evaluating vessel diameter.