University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Frankfurt, Germany.
University Hospital Frankfurt, Department of Diagnostic and Interventional Radiology, Frankfurt, Germany.
Eur J Radiol. 2018 Feb;99:111-117. doi: 10.1016/j.ejrad.2017.12.024. Epub 2017 Dec 30.
To investigate the impact of traditional (VMI) and noise-optimized virtual monoenergetic imaging (VMI+) algorithms on quantitative and qualitative image quality, and the assessment of stenosis in carotid and intracranial dual-energy CTA (DE-CTA).
DE-CTA studies of 40 patients performed on a third-generation 192-slice dual-source CT scanner were included in this retrospective study. 120-kVp image-equivalent linearly-blended, VMI and VMI+ series were reconstructed. Quantitative analysis included evaluation of contrast-to-noise ratios (CNR) of the aorta, common carotid artery, internal carotid artery, middle cerebral artery, and basilar artery. VMI and VMI+ with highest CNR, and linearly-blended series were rated qualitatively. Three radiologists assessed artefacts and suitability for evaluation at shoulder height, carotid bifurcation, siphon, and intracranial using 5-point Likert scales. Detection and grading of stenosis were performed at carotid bifurcation and siphon.
Highest CNR values were observed for 40-keV VMI+ compared to 65-keV VMI and linearly-blended images (P < 0.001). Artefacts were low in all qualitatively assessed series with excellent suitability for supraaortic artery evaluation at shoulder and bifurcation height. Suitability was significantly higher in VMI+ and VMI compared to linearly-blended images for intracranial and ICA assessment (P < 0.002). VMI and VMI+ showed excellent accordance for detection and grading of stenosis at carotid bifurcation and siphon with no differences in diagnostic performance.
40-keV VMI+ showed improved quantitative image quality compared to 65-keV VMI and linearly-blended series in supraaortic DE-CTA. VMI and VMI+ provided increased suitability for carotid and intracranial artery evaluation with excellent assessment of stenosis, but did not translate into increased diagnostic performance.
研究传统（VMI）和噪声优化的虚拟单能量成像（VMI+）算法对定量和定性图像质量的影响，并评估颈内和颅内双能量 CT 血管造影（DE-CTA）中的狭窄程度。
本回顾性研究纳入了在第三代 192 层双源 CT 扫描仪上进行的 40 例 DE-CTA 研究。采用 120kVp 图像等效线性混合、VMI 和 VMI+系列进行重建。定量分析包括评估主动脉、颈总动脉、颈内动脉、大脑中动脉和基底动脉的对比噪声比（CNR）。评估 VMI 和 VMI+中 CNR 最高的序列，并对线性混合序列进行定性评估。三位放射科医生使用 5 分李克特量表评估肩部、颈动脉分叉处、虹吸部和颅内的伪影和评估的适宜性。在颈动脉分叉处和虹吸部进行狭窄的检测和分级。
与 65keV 的 VMI 和线性混合图像相比，40keV 的 VMI+显示出最高的 CNR 值（P<0.001）。所有定性评估序列中的伪影均较低，在上部主动脉的评估中具有极佳的适用性。与线性混合图像相比，VMI+和 VMI 序列在颅内和颈内动脉评估中的适用性显著更高（P<0.002）。VMI 和 VMI+在颈动脉分叉和虹吸部的狭窄检测和分级方面具有极好的一致性，且诊断性能无差异。
与 65keV 的 VMI 和线性混合序列相比，在上部主动脉的 DE-CTA 中，40keV 的 VMI+显示出改善的定量图像质量。VMI 和 VMI+提高了对颈动脉和颅内动脉的评估的适用性，具有极好的狭窄评估能力，但并未提高诊断性能。
