Quality of virtual-non-contrast phases derived from arterial and delayed phases of fast-kVp switching dual-energy CT in patients after endovascular aortic repair.
Wojciech KazimierczakNatalia KazimierczakZbigniew SerafinPublished in: The international journal of cardiovascular imaging (2023)
Objective of this study is: to analyze CT numbers in arteries and endoleaks in true non-contrast (TNC) and virtual non-contrast phases derived from arterial (VNCa) and delayed (VNCd) phases of dual-energy CT (DECT) in patients after endovascular aneurysm repair (EVAR); to assess the impact of image noise on subjective image quality parameters and the degree of subtraction of calcifications; to calculate effective dose (ED) reduction following replacement of TNC with VNC. The study included 97 patients after EVAR procedure. An initial single-energy TNC acquisition was followed by two DECT acquisitions. CT numbers of TNC, VNCa, VNCd were analyzed statistically. VNCd images were assessed qualitatively. The mean densities in endoleaks were 46.19 HU in TNC, 51.24 HU in VNCa, 42.24 HU in VNCd. The differences between them were statistically significant (p < 0.05). The mean signal-to-noise ratio (SNR) measured in the aorta and endoleaks was highest in VNCa, lowest in TNC images. No correlation between image noise, the results of qualitative analysis of VNCd, and the degree of subtraction of calcifications was found. Omitting TNC led to mean 6.54 ± 1.63 (SD) mSv (23.28% of total examination) ED reduction. VNC images have a higher SNR compared to TNC images with significant differences in the CT numbers between the TNC and VNC reconstructions. Image noise has no impact on the subjective image quality and the degree of subtraction of calcifications in VNCd images. The findings show a high diagnostic value of VNC images and suggest that VNCd images are optimal in the assessment of endoleaks with possible substantial ED reduction.
Keyphrases
- dual energy
- image quality
- computed tomography
- deep learning
- contrast enhanced
- convolutional neural network
- end stage renal disease
- emergency department
- optical coherence tomography
- positron emission tomography
- ejection fraction
- newly diagnosed
- magnetic resonance imaging
- magnetic resonance
- air pollution
- chronic kidney disease
- prognostic factors
- systematic review
- minimally invasive
- sleep quality