Secondary Logo

Institutional members access full text with Ovid®

Improvement of Image Quality in Unenhanced Dual-Layer CT of the Head Using Virtual Monoenergetic Images Compared With Polyenergetic Single-Energy CT

Neuhaus, Victor MD; Abdullayev, Nuran MD; Große Hokamp, Nils MD; Pahn, Gregor; Kabbasch, Christoph MD; Mpotsaris, Anastasios MD; Maintz, David MD; Borggrefe, Jan MD

doi: 10.1097/RLI.0000000000000367
Original Articles

Objectives The aims of this study were to compare virtual monoenergetic images and polyenergetic images reconstructed from unenhanced dual-layer detector computed tomography (DLCT) of the head and to determine kiloelectron volt levels that optimize image quality, particularly the gray-white matter contrast, and reduce beam hardening artifacts caused by the skull.

Materials and Methods Institutional review board approval was obtained. Forty patients that received DLCT were included in this retrospective study; of these patients, 22 were women and 18 were men. The average age was 61.5 ± 14.3 years. Virtual monoenergetic images were reconstructed from spectral base images at 40 keV to 120 keV. To calculate signal-to-noise ratio and contrast-to-noise ratio, attenuation and standard deviation of supratentorial gray and white matter were measured in virtual monoenergetic and polyenergetic images. Beam hardening artifacts were detected close to the calvarium and in the posterior fossa. Two radiologists rated the assessment of gray-white matter differentiation and of the subcalvarial space, as well as the artifacts caused by the skull and image noise. Student t test and Wilcoxon test were used to determine significance.

Results Compared with polyenergetic images, superior signal-to-noise ratio and superior contrast-to-noise ratio of gray and white matter were observed in virtual monoenergetic images at low kiloelectron volt levels (P < 0.0001). Subcalvarial artifacts were significantly lower at 120 keV (P < 0.02). Artifacts measured in the posterior fossa were generally lower at high kiloelectron volt levels; however, no statistical significance was detected. Virtual monoenergetic images were rated superior to polyenergetic images in regard to all 4 criteria (P < 0.0001). The observers reported an optimal radiological assessment of gray-white matter differentiation at 65 keV and optimal assessment of subcalvarial space at 120 keV.

Conclusions In comparison to polyenergetic images, virtual monoenergetic images reconstructed from unenhanced DLCT of the head at 65 keV and 120 keV allow to optimize gray-white matter contrast and reduce beam hardening artifacts caused by the skull, respectively.

From the Institute of Diagnostic and Interventional Radiology, University Hospital Cologne, Cologne, Germany.

Received for publication December 6, 2016; and accepted for publication, after revision, February 7, 2017.

Conflicts of interest and sources of funding: none declared.

Correspondence to: Victor Neuhaus, MD, Institute of Diagnostic and Interventional Radiology, University Hospital Cologne, Kerpener Straße 62, 50937 Cologne, Germany. E-mail:

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.