Virtual Unenhanced CT Research Articles

Simultaneous vessel segmentation and unenhanced prediction using self-supervised dual-task learning in 3D CTA (SVSUP)

Background and objective: The vessel segmentation in CT angiography (CTA) provides an important basis for automatic diagnosis and hemodynamics analysis. Virtual unenhanced (VU) CT images obtained by dual-energy CT can assist clinical diagnosis and reduce radiation dose by obviating true unenhanced imaging (UECT). However, accurate segmentation of all vessels in the head-neck CTA (HNCTA) remains a challenge, and VU images are currently not available from conventional single-energy CT imaging. Methods:In this paper, we proposed a self-supervised dual-task deep learning strategy to fully automatically segment all vessels and predict unenhanced CT images from single-energy HNCTA based on a developed iterative residual-sharing scheme. The underlying idea was to use the correlation between the two tasks to improve task performance while avoiding manual annotation for model training. Results: The feasibility of the strategy was verified using the data of 24 patients. For vessel segmentation task, the proposed model achieves a significantly higher average Dice coefficient (84.83%, P-values 10−3 in paired t-test) than the state-of-the-art segmentation model, vanilla VNet (78.94%), and several popular 3D vessel segmentation models, including Hessian-matrix based filter (62.59%), optically-oriented flux (66.33%), spherical flux model (66.91%), and deep vessel net (66.47%). For the unenhanced prediction task, the average ROI-based error compared to the UECT in the artery tissue is 6.1±4.5 HU, similar to previously reported 6.4±5.1 HU for VU reconstruction. Conclusions: Results show that the proposed dual-task framework can effectively improve the accuracy of vessel segmentation in HNCTA, and it is feasible to predict the unenhanced image from single-energy CTA, providing a potential new approach for radiation dose saving. Moreover, to our best knowledge, this is the first reported annotation-free deep learning-based full-image vessel segmentation for HNCTA.

Virtual Unenhanced Dual-Energy CT Images Obtained with a Multimaterial Decomposition Algorithm: Diagnostic Value for Renal Mass and Urinary Stone Evaluation.

Background Virtual unenhanced (VUE) images obtained by using a dual-energy CT (DECT) multimaterial decomposition algorithm hold promise for diagnostic use in the abdomen in lieu of true unenhanced (TUE) images. Purpose To assess VUE images obtained from a DECT multimaterial decomposition algorithm in patients undergoing renal mass and urinary stone evaluation. Materials and Methods In this retrospective Health Insurance Portability and Accountability Act-compliant study, DECT was performed in patients undergoing evaluation for renal mass or urinary stone. VUE images were compared quantitatively to TUE images and qualitatively assessed by four independent radiologists. Differences in attenuation between VUE and TUE images were summarized by using 95% limits of agreement. Diagnostic performance in urinary stone detection was summarized by using area under the receiver operating characteristic curve, sensitivity, and specificity. Results A total of 221 patients (mean age ± standard deviation, 61 years ± 14; 129 men) with 273 renal masses were evaluated. Differences in renal mass attenuation between VUE and TUE images were within 3 HU for both enhancing masses (95% limits of agreement, -3.1 HU to 2.7 HU) and nonenhancing cysts (95% limits of agreement, -2.9 HU to 2.5 HU). Renal mass classification as enhancing mass versus nonenhancing cyst did not change (reclassification rate of enhancing masses, 0% [0 of 78]; 95% CI: 0, 5; reclassification rate of nonenhancing cysts, 0% [0 of 193]; 95% CI: 0, 2) with use of VUE in lieu of TUE images. Among 166 urinary stones evaluated, diagnostic performance of VUE images for stone detection was lower compared with that of TUE images (area under the receiver operating characteristic curve, 0.79 [95% CI: 0.73, 0.84] vs 0.93 [95% CI: 0.91, 0.95]; P < .001) due to reduced sensitivity of VUE for detection of stones 3 mm in diameter or less compared with those greater than 3 mm (sensitivity, 23% [25 of 108; 95% CI: 12, 40] vs 88% [126 of 144; 95% CI: 77, 94]; P < .001). Conclusion Compared with true unenhanced images, virtual unenhanced (VUE) images were unlikely to change renal mass classification as enhancing mass versus nonenhancing cyst. Diagnostic performance of VUE images remained suboptimal for urinary stone detection due to subtraction of stones 3 mm or less in diameter. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Sosna in this issue.

Visualization of joint and bone using dual-energy CT arthrography with contrast subtraction: in vitro feasibility study using porcine joints

To assess the feasibility of subtracting various concentrations of iodinated contrast material on dual-energy computed tomographic (DECT) arthrography to provide both CT arthrography and virtual unenhanced CT (VUCT) in a single CT acquisition. This was an in vitro study for which institutional review board approval was not required. CT arthrographies of 12 joints of pig cadavers were obtained using dual-energy CT. Various concentrations of iodinated contrast material, 25% (75mg/ml), 50% (150mg/ml), 75% (225mg/ml), and 100% (300mg/ml) were used for the DECT arthrography. The paired regions of interest (ROI) were drawn over the same location on two paired CT scans at different tube voltages (80 kVp and 140 kVp). The average Hounsfield units (HU) ratio of the contrast media(HU on CT at 80 kVp/HU on CT at 140kVp) was calculated for each joint. Subtraction of contrast material was carried out using VUCT application The 25% iodinated contrast mixture was successfully subtracted from DECT arthrography of four joints, in which the average HU ratio ranged from 1.95 to 2.0. The subtraction of the 50%,75%, and 100% iodine contrast mixtures was not successful, because of the upper demonstrable HU limit in dual-energy CT. DECT arthrography with 25% iodinated contrast medium injection can provide both CT arthrography and virtual unenhanced images in a single CT acquisition.

Attenuation Correction Maps for SPECT Myocardial Perfusion Imaging from Contrast-Enhanced Coronary CT Angiography: Gemstone Spectral Imaging with Single-Source Dual Energy and Material Decomposition

The aim of this study was to explore the feasibility of attenuation correction (AC) of myocardial perfusion imaging (MPI) with a virtual unenhanced cardiac CT scan synthesized from contrast-enhanced single-source dual-energy coronary CT angiography. Segmental myocardial percentage uptake values obtained with AC were analyzed by use of correlation analysis and Bland-Altman limits of agreement (20-segment model), and clinical agreement was evaluated in 30 patients. The 2 methods showed an excellent correlation for segmental myocardial percentage uptake at stress (r = 0.93; P < 0.001; low dose) and at rest (r = 0.90; P < 0.001; high dose) with narrow Bland-Altman limits of agreement (-6.8% to 7.8% and -7.8% to 7.4%, respectively). The levels of clinical agreement of SPECT MPI corrected with standard versus virtual unenhanced CT AC were 99% per coronary territory and 97% per patient. Our results suggest that AC of SPECT MPI with a virtual unenhanced CT scan synthesized from contrast-enhanced coronary CT angiography is feasible and reliable.

Evaluation of virtual unenhanced CT obtained from dual-energy CT urography for detecting urinary stones

The aim of our study was to determine if virtual unenhanced CT (VUCT) is equivalent to unenhanced CT (UCT) for detecting urinary stones. Our institutional review board approved this retrospective study, which was compliant with the Health Insurance Portability and Accountability Act. A total of 80 stones were detected in 32 patients among 146 consecutive patients undergoing dual-energy CT urography. The number and size of stones were recorded on nephrographic VUCT (NVUCT) and excretory VUCT (EVUCT) images, respectively. UCT was a reference of standard for the number and size of stones. Image quality of VUCT was qualitatively assessed using a five-point scale. Repeated-measures analysis of variance with post-test was used for statistical analysis. 62 stones in 29 patients were detected on NVUCT and 59 stones in 27 patients were detected on EVUCT. The size of stones detected on NVUCT or EVUCT was significantly smaller compared with stones on UCT (p<0.05). The size of stones detected on UCT, NVUCT and EVUCT ranged from 1.4 to 19.2 mm (mean, 4.6 mm), 0 to 19.2 mm (mean, 3.6 mm) and 0 to 18.7 mm (mean, 3.6 mm), respectively. 18 stones were missed on NVUCT and 21 were missed on EVUCT. The sizes ranged from 1.4 to 3.2 mm (mean, 2.1 mm) and 1.4 to 3.2 mm (mean, 2.2 mm) on UCT, respectively. VUCT was inferior to UCT regarding image quality (p<0.05). VUCT missed a significant number of small stones probably owing to poor image quality compared with UCT. Subsequently, VUCT cannot replace UCT for detecting urinary stones.

Virtual unenhanced second generation dual-source CT of the liver: Is it time to discard the conventional unenhanced phase?

Dual-energy dual source CT can almost simultaneously image patients using two different tube potentials, allowing material decomposition and creation of 'virtual unenhanced' (VU) images from post-contrast series. 75 patients undergoing triple-phase liver CT examinations were imaged using a second generation dual-source CT machine with tube potentials 140/100 kVp. Post-processing VU series were derived from arterial and portal phases. Regions-of-interest from liver parenchyma and within fat ('noise' assessment) were drawn to compare VU series to conventional unenhanced (CU) series. Subjective analysis assessed image quality and the suitability of VU to replace CU series. Mean Hounsfield unit (HU) values of liver were higher in the VU series: portal 51.9 (SD = 10.29), arterial 51.1 (SD = 10.05), compared to the CU series 49.2 (SD = 9.11); P<0.001. However, Pearson's correlation of the VU and CU series remained excellent: 0.838 (portal), 0.831 (arterial). Bland-Altman plots also showed good agreement between both VU and the CU datasets. Noise measurements were significantly lower in both VU series (P<0.001). For subjective analysis, image quality was rated as very good/excellent in 100% of CU images, 93.3% of portal VU and 88.7% of arterial VU series. Overall, portal VU and arterial VU images were acceptable replacements for the CU series in 97.4% and 96.1%, respectively. Post-processing was noted to create a number of artefacts in VU images--knowledge of these is essential for interpretation. Portal and arterial-derived VU images objectively correlate to CU images and demonstrate good image quality and acceptability. VU image sets could replace the conventional unenhanced images in the vast majority of cases, significantly reducing radiation dose.

Dual Energy CT for Monitoring Targeted Therapies in Patients with Advanced Gastrointestinal Stromal Tumor: Initial Results

Advanced gastrointestinal stromal tumours (GISTs) are treated with tyrosine kinase inhibitors, which also have antiangiogenic properties. Dual-energy CT (DECT) allows to acquire semi-quantitative iodine images which might correlate with blood pool and tumor vascularity. In this feasibility-study, we correlated lesional iodine uptake estimations in correlation to tumor size changes under targeted therapy as first step in the evaluation of dedicated DECT based strategies for monitoring molecular therapies in GIST. 48 tumor lesions in 18 patients with metastasized histologically proven GIST under tyrosine kinase inhibitor (TKI) therapy were analyzed. Patients were examined with a dual-source CT in dual-energy mode (Voltage tube A: 80 kV, tube B: 140 kV). Using the dual-energy software virtual unenhanced, selective iodine (overlay) and mixed CT numbers (similar to CT numbers at 120 kV) of lesions were calculated. The largest diameter of each lesion on cross-sectional axial images was measured. The mean difference of overlay CT numbers in the baseline and follow-up examinations was calculated and this marker of lesional iodine uptake was compared to lesional size changes under molecular therapy. Utilizing the cut-off value 15 HU of change in overlay, DECT allowed to identify lesions with a stable, increased or decreased lesional iodine uptake with corresponding typical lesion size change patterns after 3 months of targeted therapy: 30 lesions had no significant change of overlay CT numbers (OL) (mean: -2.4 HU) or lesion size (mean: +1.5%). A strong decline of the OL (mean: - 24 HU) in 13 lesions was combined with a pronounced growth (mean: + 26%). 5 lesions showed a strong increase of the absolute OL (mean: + 23 HU) associated with a moderate increase in size (+ 8%). Determination of the overlay CT number with DECT enables to stratify metastases with stable, increasing or decreasing iodine uptake over time with -in our collective- typical lesion size change patterns. Investigation of a larger patient cohort, comparison to histology, alternate imaging biomarkers and correlatrion to long-term response will further clarify the significance of these findings for monitoring targeted therapies in GIST.

Feasibility of Dual-Energy CT in the Arterial Phase: Imaging After Endovascular Aortic Repair

The purpose of this study was to investigate replacing unenhanced and arterial single-energy CT acquisitions after endovascular aneurysm repair with one dual-energy CT arterial acquisition. Thirty patients underwent arterial dual-energy CT (80 and 140 kVp) and venous single-energy CT (120 kVp) after endovascular aneurysm repair, and the radiation doses were compared with those of a standard triple-phase protocol. Both virtual unenhanced and arterial images were generated with dual-energy CT. Images were reviewed clinically for detection of endoleaks and evaluation of stent and calcium appearance. The aortic luminal attenuation on virtual unenhanced CT images was compared with that on previously acquired true unenhanced images. Virtual unenhanced, arterial, and venous images were compared for thrombus attenuation. Single-energy CT and dual-energy CT images were compared for noise. Replacement of two (unenhanced, arterial) of three single-energy CT acquisitions with one dual-energy CT acquisition resulted in 31% radiation dose savings. All images were clinically interpretable. Thoracic (32 +/- 2 vs 35 +/- 4 HU) and abdominal (30 +/- 3 vs 35 +/- 5 HU) aortic attenuation was similar on virtual unenhanced and true unenhanced images. Thrombus attenuation was similar on virtual unenhanced (32 +/- 6 HU), arterial phase (33 +/- 7 HU), and venous phase (34 +/- 6 HU) images. Decreased stent and calcium attenuation was observed at some locations on virtual unenhanced images. Noise in the thoracic (10 +/- 1 HU) and abdominal (12 +/- 2 HU) aorta was lower on virtual unenhanced images than on true unenhanced images (13 +/- 4 HU, 19 +/- 5 HU). Noise was comparable for dual-energy and single-energy CT (thorax, 16 +/- 2 vs 13 +/- 2 HU; abdomen, 21 +/- 3 vs 23 +/- 5 HU). Virtual unenhanced and arterial phase images derived from dual-energy CT can replace true unenhanced and arterial phase single-energy CT images in follow-up after endovascular aneurysm repair (except immediately after the procedure), providing comparable diagnostic information with substantial dose savings.