Interscanner and Intrascanner Comparison of Virtual Unenhanced Attenuation Values Derived From Twin Beam Dual-Energy and Dual-Source, Dual-Energy Computed Tomography.

Published

Journal Article

OBJECTIVE: The aim of the current study was to evaluate the reliability and comparability of virtual unenhanced (VUE) attenuation values derived from scans of a single-source, dual-energy computed tomography using a split-filter (tbDECT) to a dual-source dual-energy CT (dsDECT). MATERIALS AND METHODS: In this retrospective study, comparisons for tbDECT and dsDECT were made within and between different dual-energy platforms. For the interscanner comparison, 126 patients were scanned with both scanners within a time interval of 224 ± 180 days; for the intrascanner comparison, another 90 patients were scanned twice with the same scanner within a time interval of 136 ± 140 days. Virtual unenhanced images were processed off of venous phase series. Attenuation values of 7 different tissues were recorded. Disagreement for VUE HU measurements greater than 10 HU between 2 scans was defined as inadequate. RESULTS: The interscanner analysis showed significant difference between tbDE and dsDE VUE CT values (P < 0.01) for 6 of 7 organs. Percentage of cases that had more than 10 HU difference between tbDE and dsDE for an individual patient ranged between 15% (left kidney) and 62% (spleen).The intrascanner analysis showed no significant difference between repeat scans for both tbDECT and dsDECT (P > 0.05). However, intrascanner disagreements for the VUE HU measurements greater than 10 HU were recorded in 10% of patients scanned on the tbDECT and 0% of patients scanned on the dsDECT. The organs with the highest portion of greater than 10 HU errors were the liver and the aorta (both 20%). CONCLUSIONS: Dual-energy techniques vary in reproducibility of VUE attenuation values. In the current study, tbDECT demonstrated higher variation in VUE HU measurements in comparison to a dsDECT. Virtual unenhanced HU measurements cannot be reliably compared on follow-up CT, if these 2 different dual-energy CT platforms are used.

Full Text

Duke Authors

Cited Authors

  • Obmann, MM; Kelsch, V; Cosentino, A; Hofmann, V; Boll, DT; Benz, MR

Published Date

  • January 2019

Published In

Volume / Issue

  • 54 / 1

Start / End Page

  • 1 - 6

PubMed ID

  • 30096063

Pubmed Central ID

  • 30096063

Electronic International Standard Serial Number (EISSN)

  • 1536-0210

Digital Object Identifier (DOI)

  • 10.1097/RLI.0000000000000501

Language

  • eng

Conference Location

  • United States