TH-E-217BCD-04: MA Modulation and Iterative Reconstruction: Evaluation Using a New CT Phantom

Abstract

Purpose: The routine clinical CT use of tube current modulation (TCM) and iterative reconstruction algorithms, which can be used to reduce radiation dose, necessitates evaluating their effects on image quality. This study's purpose was to use a new phantom (Mercury Phantom) to evaluate the quality of images acquired with TCM and iterative reconstruction algorithms as implemented by two different CT manufacturers. Methods: The phantom consists of four cylindrical and three tapered regions. The phantom was designed to measure noise-power spectrum (NPS) and modulation transfer function (MTF) across a range of diameters'” 16, 23, 30, and 37 cm'”and at multiple, clinically relevant insert materials from air to iodinated tissue. Images were acquired on a Siemens Flash and GE 750HD using a range of TCM and proprietary reconstruction algorithms: ASIR, IRIS, and SAFIRE. Custom software measured the non-stationarity of NPS as a function of reconstruction and TCM. Contrast and MTF were measured for each insert for each reconstruction algorithm. Results: Measured HU of different materials varied minimally as a function of reconstruction algorithm and dose reduction technique. However, as a function of phantom size from the 16-cm to 37-cm diameter, the [air,Polystyrene,Acrylic,Teflon] HU of 120-kVp images changed by approximately [50,−20,−8,−130] for the 750HD and [35,0,6,−75] for the Flash. MTFs of the linear reconstruction algorithms were independent of size, TCM, and contrast, but for iterative algorithms, the MTFs depended on the object contrast and improved by as much as 25%. Peak NPS decreased from 37-cm to 16-cm (∼86%), decreased for iterative reconstruction relative to linear algorithms (∼21%), and decreased radially from the phantom's center (∼25%). mA modulation performance differed considerably between the two systems. Conclusion: The Mercury Phantom can be used to assess trade-offs in CT image quality and dose reduction techniques for a range of sizes incorporating TCM and iterative reconstruction. Joshua M. Wilson Funded, in part, by Research Agreement with GE Healthcare. Ehsan Samei Research grant: NIH R01 EB001838; Research grant: General Electric; Research grant: Siemens Medical; Research grant: Carestream Health; Share-holder: Zumatek Inc; Advisory Board: IBA