Premium
A comparison of digital radiography systems in terms of effective detective quantum efficiency
Author(s) -
Bertolini Marco,
Nitrosi Andrea,
Rivetti Stefano,
Lanconelli Nico,
Pattacini Pierpaolo,
Ginocchi Vladimiro,
Iori Mauro
Publication year - 2012
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4704500
Subject(s) - detective quantum efficiency , imaging phantom , optical transfer function , optics , computed radiography , attenuation , digital radiography , magnification , physics , quantum noise , noise (video) , radiography , medical imaging , image quality , attenuation coefficient , nuclear medicine , radiometry , computer science , medicine , quantum , artificial intelligence , quantum mechanics , nuclear physics , image (mathematics)
Purpose: The purpose of this study is to compare digital radiography systems using the metric effective detective quantum efficiency (eDQE), which better reflects digital radiography imaging system performance under clinical operating conditions, in comparison with conventional metrics such as modulation transfer function (MTF), normalized noise power spectra (NNPS), and detective quantum efficiency (DQE).Methods: The eDQE was computed by the calculation of the MTF, the NNPS, the phantom attenuation and scatter, and estimation of x‐ray flux. The physical characterization of the systems was obtained with the standard beam conditions RQA5 and RQA9, using the PA Chest phantom proposed by AAPM Report # 31 simulating the attenuation and scatter characteristics of the adult human thorax. The MTF (eMTF) was measured by using an edge test placed at the frontal surface of the phantom, the NNPS (eNNPS) was calculated from images of the phantom acquired at three different exposure levels covering the operating range of the system (E 0 , which is the exposure at which a system is normally operated, 1/3 E 0 , and 3 E0), and scatter measurements were assessed by using a beam‐stop technique. The integral of DQE (IDQE) and eDQE (IeDQE) was calculated over the whole spatial frequency range.Results: The eMTF results demonstrate degradation due to magnification and the presence of scattered radiation. The eNNPS was influenced by the grid presence, and in some systems, it contained structured noise. At typical clinical exposure levels, the magnitude of eDQE(0) with respect to DQE(0) at RQA9 beam conditions was 13%, 17%, 16%, 36%, and 24%, respectively, for Carestream DRX‐1, Carestream DRX‐1C, Carestream Direct View CR975, Philips Digital Diagnost VM, and GE Revolution XR/d. These results were confirmed by the ratio of IeDQE and IDQE in the same conditions.Conclusions: The authors confirm the robustness and reproducibility of the eDQE method. As expected, the DR systems performed better than the CR systems due to their superior signal‐to‐noise transfer characteristics. The results of this study suggest the eDQE method may provide an opportunity to more accurately assess the clinical performance of digital radiographic imaging systems by accounting for factors such as the presence of scatter, use of an antiscatter grid, and magnification and focal spot blurring effects, which are not reflected in conventional DQE measures.