Hybrid spectral CT reconstruction

Clark, DP; Badea, Cristian T.

doi:10.1371/journal.pone.0180324

Cited by 45 publications

(68 citation statements)

References 69 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our data-adaptive iterative reconstruction (B) roughly equalizes the reconstruction error across all time points and energies, despite a nearly 5-fold difference in reconstruction errors between the lowest and highest energy threshold following algebraic reconstruction (A). The accuracy of these reconstruction results is consistent with our previous, detailed analysis of spectral bias introduced by RSKR and our iterative reconstruction approach [9].…”

Section: Resultssupporting

confidence: 88%

“…Following from our past work on data-adaptive, iterative reconstruction methods for temporal and spectral CT [9, 18, 25], Fig 1 summarizes our approach to these multi-channel reconstruction problems. We solve a series of weighted algebraic reconstruction problems (columns of X, channels indexed by c ), subject to relaxed constraints which ensure consistency between channels (“Reg”; scalar regularization parameter, λ), using the split Bregman method with the add-residual-back strategy [16] and a low-rank and sparse signal model [17].…”

Section: Methodsmentioning

confidence: 99%

“…We initially proposed this framework for 4D, DE CT reconstruction using EIDs [18]. We later extended this framework for spectral CT reconstruction with an arbitrary number of spectral channels, proposing a regularizer called rank-sparse kernel regression (RSKR) [9]. In this paper, we combine our original 4D, DE reconstruction framework with RSKR, enabling in vivo , 4D PCD-CT reconstruction.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Photon-counting cine-cardiac CT in the mouse

Clark

Holbrook

Lee

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

12The maturation of photon-counting detector (PCD) technology promises to 13 enhance routine CT imaging applications with high-fidelity spectral information. In this 14 paper, we demonstrate the power of this synergy and our complementary reconstruction 15 techniques, performing 4D, cardiac PCD-CT data acquisition and reconstruction in a 16 mouse model of atherosclerosis, including calcified plaque. Specifically, in vivo cardiac 17 micro-CT scans were performed in four ApoE knockout mice, following their development 18 of calcified plaques. The scans were performed with a prototype PCD (DECTRIS, Ltd.) 19 with 4 energy thresholds. Projection sampling was performed with 10 ms temporal 20 resolution, allowing the reconstruction of 10 cardiac phases at each of 4 energies (40, 3D 21 volumes per mouse scan). Reconstruction was performed iteratively using the split 34 detected during slow moving phases of the cardiac cycle. Given these preliminary results, 35 we believe that PCD technology will enhance dynamic CT imaging applications with high-36 fidelity spectral and material information. 37 42 Philips (dual-layer). These DE scanners are associated with several routine imaging 43 applications, including plaque differentiation, myocardial perfusion, and kidney stone 44 characterization [5]. Preclinically, spectral CT has been utilized in several additional 3 45 applications, including differential imaging of vasculature and vascular permeability in 46 sarcoma [6] and lung [7] tumors, and with several preclinical contrast agents based on 100 gene [22], show a marked increase in total plasma cholesterol levels, and are prone to 101 develop atherosclerotic lesions [23]. At 8-12 weeks of age, the ApoE-/-mice were 102 exposed to 25 fractions of 2 Gy partial-heart irradiation, mimicking cardiac radiation 103 exposure of breast cancer patients treated with radiation therapy [24]. After irradiation, 104 the mice were kept on a regular diet for a year prior to imaging. Three days before 105 imaging, the mice were intravenously injected with gold nanoparticle contrast agent (15 106 nm AuroVist, www.nanoprobes.com) at a dose of 0.004 mL/g mouse, with gold 107 accumulation expected at the site of any myocardial injury [24]. Notably, with an 80 kVp 108 source spectrum, K-edge imaging of gold (edge at 80.7 keV) was not possible. Three 109 days later, immediately prior to imaging, the same mice were injected with a liposomal 110 iodinated contrast agent (described in [8]; 0.012 mL/g mouse) to measure cardiac 111 functional metrics. The free-breathing mice were scanned while under anesthesia 112 induced with 1-2% isoflurane delivered by nose cone. Both ECG and respiratory signals 6 113 were recorded during scanning (average heart rate: 448 ± 50 bpm; average respiratory 114 rate: 142 ± 19 breaths/min.). The ECG signal was used for retrospective projection 115 weighting during reconstruction. Respiratory gating was not performed. 116 Multi-channel image reconstruction 117Following from our past work on data-adaptive, iterative reconstruc...

show abstract

Section: Resultssupporting

confidence: 88%

Section: Methodsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Photon-counting cine-cardiac CT in the mouse

Clark

Holbrook

Lee

et al. 2019

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Unlike DECT, SPCCT uses energy-resolving detectors that can simultaneously sample the energy spectrum at multiple regions. This allows a higher spectral resolution, enabling the identification of material-specific spectral characteristics, such as the K-edge signature of the contrast agent [ 5 , 15 – 18 ]. This is particularly interesting as the K-edge energies of contrast agents that contain heavy elements such as gadolinium, ytterbium, bismuth or gold are within the clinical x-ray tube spectrum [ 1 , 17 , 19 – 21 ].…”

Section: Introductionmentioning

confidence: 99%

Multicolour imaging with spectral photon-counting CT: a phantom study

Si‐Mohamed¹,

Bar-Ness²,

Sigovan³

et al. 2018

Eur Radiol Exp

View full text Add to dashboard Cite

BackgroundTo evaluate the feasibility of multicolour quantitative imaging with spectral photon-counting computed tomography (SPCCT) of different mixed contrast agents.MethodsPhantoms containing eleven tubes with mixtures of varying proportions of two contrast agents (i.e. two selected from gadolinium, iodine or gold nanoparticles) were prepared so that the attenuation of each tube was about 280 HU. Scans were acquired at 120 kVp and 100 mAs using a five-bin preclinical SPCCT prototype, generating conventional, water, iodine, gadolinium and gold images. The correlation between prepared and measured concentrations was assessed using linear regression. The cross-contamination was measured for each material as the root mean square error (RMSE) of its concentration in the other material images, where no signal was expected. The contrast-to-noise ratio (CNR) relative to a phosphate buffered saline tube was calculated for each contrast agent.ResultsThe solutions had similar attenuations (279 ± 10 HU, mean ± standard deviation) and could not be differentiated on conventional images. However, a distinction was observed in the material images within the same samples, and the measured and prepared concentrations were strongly correlated (R2 ≥ 0.97, 0.81 ≤ slope ≤ 0.95, -0.68 ≤ offset ≤ 0.89 mg/mL). Cross-contamination in the iodine images for the mixture of gold and gadolinium contrast agents (RMSE = 0.34 mg/mL) was observed. CNR for 1 mg/mL of contrast agent was better for the mixture of iodine and gadolinium (CNRiodine = 3.20, CNRgadolinium = 2.80) than gold and gadolinium (CNRgadolinium = 1.67, CNRgold = 1.37).ConclusionsSPCCT enables multicolour quantitative imaging. As a result, it should be possible to perform imaging of multiple uptake phases of a given tissue/organ within a single scan by injecting different contrast agents sequentially.

show abstract

“…Several companies brought to the market so called spectral CT devices. They are usually based on dual energy X-ray sources, and the comparison of the energy-dependent attenuation of signal can improve the soft tissue recognition and distinguish contrasts and the bone, which is not possible in classical CT devices [155]. These CT machines can utilize slightly lower radiation dose compared to previous generation.…”

Section: Preclinical Hybrid Imagingmentioning

confidence: 99%