Back to Search Start Over

Poly-energetic and virtual mono-energetic images from a novel dual-layer spectral detector CT: optimization of window settings is crucial to improve subjective image quality in abdominal CT angiographies.

Authors :
Doerner J
Luetkens JA
Iuga AI
Byrtus J
Haneder S
Maintz D
Hickethier T
Source :
Abdominal radiology (New York) [Abdom Radiol (NY)] 2018 Mar; Vol. 43 (3), pp. 742-750.
Publication Year :
2018

Abstract

Purpose: We aimed to determine optimal window settings for poly-energetic (PolyE) and virtual mono-energetic images (MonoE) derived from abdominal angiographic studies on a novel dual-layer spectral detector CT (SDCT) system.<br />Methods: From 50 patients, SDCT datasets PolyE and MonoE at 70 and 40 keV levels were reconstructed and best individual window width and level (BI-W/L) manually assessed. Through regression analysis, the so-called optimized individual (OI-W/L) values were obtained. Subjective image quality parameters and vessel diameters were measured to determine influences of different W/L settings.<br />Results: Image noise was lower and attenuation and contrast-to-noise ratio were higher in MonoE compared to PolyE (all p ≤ 0.002). Mean BI-W/L values for PolyE, 70, and 40 keV were 637/284, 647/291, and 1568/691, respectively. Mean OI-W/L values were 631/276, 628/286, and 1516/667, respectively. Compared to standard settings, all adjusted W/L settings varied significantly and yielded higher subjective scoring. No between-group differences were found between manually adjusted and mathematically calculated W/L settings.<br />Conclusion: PolyE and MonoE from abdominal angiographic SDCT studies require appropriate W/L settings especially at low energy reconstruction levels. Individual adjustment reaches the best image quality but is time consuming. From our data, predefined W/L settings of 640/280 (PolyE/MonoE 70 keV) and 1570/690 (MonoE 40 keV) as a non-individualized starting point for abdominal angiographic studies from the novel SDCT system are suggested.

Details

Language :
English
ISSN :
2366-0058
Volume :
43
Issue :
3
Database :
MEDLINE
Journal :
Abdominal radiology (New York)
Publication Type :
Academic Journal
Accession number :
28677003
Full Text :
https://doi.org/10.1007/s00261-017-1241-1