Single-shot Dual-energy Imaging Research Articles

Design and fabrication of a sandwich detector for material discrimination and contrast cancellation in dual-energy based x-ray imaging

Dual-energy imaging represents a versatile and evolving technology with wide-ranging applications in medicine and beyond. Recent technological developments increased the potential for improved diagnostic accuracy and expanded imaging capabilities across various fields. The purpose of this work is to design and develop an energy-integrating multilayer detector, known as a sandwich detector, aimed at single-shot dual-energy imaging tasks such as material discrimination and contrast cancellation. The sandwich detector uses two complementary metal oxide semiconductor advanced pixel sensors of 50 μm pixel size. The top and bottom sensors detect low-energy (LE) and high-energy (HE) photons, with sensors coupled with 250 and 600 μm scintillators, respectively. For better spectral separation between layers without excessively affecting the detected statistic in the bottom layer, the insertion of a 0.25-mm Cu filter between the layers was found to be the optimal choice, from among the tested 0-, 0.25-, and 0.5-mm filter options. The thickness selection for scintillator and intermediate Cu filter was carried out through a dual-energy simulation model. The experiments confirmed the model's reliability in selecting the optimal thicknesses of the intermediate Cu filter, thereby providing reassurance also on the choice of the top scintillator.

Noise-reduction approaches to single-shot dual-energy imaging with a multilayer detector

The single-shot dual-energy (DE) method using a sandwich-like multilayer detector generally suffers from noise in the resultant DE images due to the poor quantum noise in images obtained from the rear detector layer and its amplification due to the subtraction operation in DE reconstruction. The use of interlayer metal filter can further increase noise in DE images by increasing the rear-detector quantum noise. We have adopted several noise-reduction algorithms to the single-shot DE reconstruction, which included the Gaussian-filtering noise reduction (GNR), the median-filtering noise reduction (MNR), and the anti-correlated noise reduction (ANR). We assessed the effectiveness of noise-reduction methods by investigating noise and dose-normalized contrast-to-noise ratio for a mouse-mimicking phantom consisting of aluminum bone and polyurethane soft tissues. Noise-power spectrum was also measured to investigate correlation noise. Overall, the ANR showed the best performance. At some extreme imaging technique conditions, such as a lower tube voltages and a larger spectral energy separation using thick copper sheets between the front and rear detector layers, the MNR outperformed the ANR. The performance of GNR was nearly similar to that of MNR. The results were well reflected into demonstration bone images obtained for a postmortem mouse. Although the ANR was effective to reduce noise in the single-shot DE imaging using the sandwich detector, multivariate optimization of ANR parameters with respect to imaging tasks and imaging techniques is remained as a future study.

Development of a large-area multilayer detector for single-shot dual-energy imaging

We have developed a sandwich-like multilayer detector capable of imaging a large field of view for the purpose of dental computed tomography. Two detectors forming the sandwich detector are based on indirect-conversion flat-panel detector technology. This development employed a relatively thin Gd2O2S:Tb scintillator for the front-detector scintillator and a relatively thick CsI:Tl for the rear one. On the other hand, the same photodiode arrays based on complementary metal-oxide-semiconductor active pixel technology were used for both detector layers. While the substrate of the rear flat-panel detector was a conventional printed-circuit board (PCB), the front detector was realized onto a flexible PCB, which is then electrically connected to a separate corresponding PCB placed under the rear detector PCB. Imaging performance of each detector layer in the sandwich detector was evaluated in terms of large-area or zero-frequency signal and noise, modulation-transfer function, noise-power spectrum, and detective quantum efficiency. Compared to previous designs, the current design showed higher signal-to-noise ratio performances for which the scintillators used were mainly responsible. Demonstration dual-energy images obtained for a postmortem mouse supported the quantitative measurements.

Linear modeling of single-shot dual-energy x-ray imaging using a sandwich detector

For single-shot dual-energy (DE) imaging, a sandwich detector typically consists of a thin front detector and a thick rear detector. Therefore, the spatial-resolution characteristics of the two detectors are different, and as a result, weighted subtraction of the corresponding two images gives rise to edge-enhancement characteristics in the resulting DE images. This is a unique characteristic of single-shot DE imaging compared to the conventional dual-shot DE imaging which uses the same detector to acquire low- and high-energy images. Using a linear-systems theory, in this paper, we show that the modulation-transfer function (MTF) of a sandwich detector is a weighted average of contributions from each MTF characteristic of two detector layers forming the sandwich detector. The MTF results obtained using the developed model are validated with those measured directly from single-shot DE images for an edge-knife phantom. Weighting larger than at least 0.5 in DE reconstruction gives an enhancement in DE MTF at mid and high spatial frequencies compared to the MTFs obtained from each detector layer. The behavior of the linear model as a function of weighting factor used for DE reconstruction is discussed in comparisons with numerical simulations.

Cascaded-systems analysis of sandwich x-ray detectors

Active sandwich-like multilayer detectors have been developed, and their potential for motion-artifact-free dual-energy x-ray imaging at a single exposure has been demonstrated in the material decomposition context. Since the sandwich detector uses the x-ray beam transmittance through the front layer, direct x-ray interaction within photodiodes in the front layer is unavoidable, and which can increase noise in the front detector images. Similar direct x-ray interaction can also occur in the rear detector layer. To obtain a better contrast performance, an additional filter layer can be placed between the two detector layers. However, this filter layer can increase adversely noise in images obtained from the rear detector layer by reducing the number of x-ray photons reaching it. A theoretical model, which can describe the signal-to-noise performance of the sandwich detector as functions of various design parameters, has been developed by using a linear cascaded-systems theory. From the cascaded-systems analysis, the direct x-ray interaction increases noise at the high spatial frequencies where the number of secondary quanta lessens. The intermediate filter layer enhances the contribution of additive electronic noise in the overall noise performance of the rear detector layer. The detailed cascaded-systems analysis on the x-ray sandwich detectors are reported in comparisons with the measured noise-power spectra and detective quantum efficiencies. The developed model will be useful for a better design and practical use of a sandwich detector for single-shot dual-energy imaging.

Effects of the energy-separation filter on the performance of each detector layer in the sandwich detector for single-shot dual-energy imaging

A novel sandwich-style single-shot detector has been built by stacking two indirect-conversion flat-panel detectors for preclinical dual-energy mouse imaging. Although this single-shot method is more immune to motion artifacts compared with the conventional dual-shot method (i.e., fast kVp switching), it may suffer from reduced image quality because of poor spectral separation between the two detectors. Spectral separation can be improved by using an intermediate filter between the two detector layers. Adversely, the filter reduces the number of x-ray photons reaching the rear detector, hence probably increasing image noise. For a better design and practical use of the sandwich detector for single-shot dual-energy imaging, imaging performances of each detector layer in the sandwich detector are investigated for various spectral-separation extents and applied tube voltages. The imaging performances include the modulation-transfer function, the Wiener noise-power spectrum, and the detective quantum efficiency. According to the experimental results, impacts of the intermediate filter on the imaging performances of each detector layer are marginal. The detailed experimental results are shown in this study.

Multiscale deformable registration for dual‐energy x‐ray imaging

Dual-energy (DE) imaging of the chest improves the conspicuity of subtle lung nodules through the removal of overlying anatomical noise. Recent work has shown double-shot DE imaging (i.e., successive acquisition of low- and high-energy projections) to provide detective quantum efficiency, spectral separation (and therefore contrast), and radiation dose superior to single-shot DE imaging configurations (e.g., with a CR cassette). However, the temporal separation between high-energy (HE) and low-energy (LE) image acquisition can result in motion artifacts in the DE images, reducing image quality and diminishing diagnostic performance. This has motivated the development of a deformable registration technique that aligns the HE image onto the LE image before DE decomposition. The algorithm reported here operates in multiple passes at progressively smaller scales and increasing resolution. The first pass addresses large-scale motion by means of mutual information optimization, while successive passes (2-4) correct misregistration at finer scales by means of normalized cross correlation. Evaluation of registration performance in 129 patients imaged using an experimental DE imaging prototype demonstrated a statistically significant improvement in image alignment. Specific to the cardiac region, the registration algorithm was found to outperform a simple cardiac-gating system designed to trigger both HE and LE exposures during diastole. Modulation transfer function (MTF) analysis reveals additional advantages in DE image quality in terms of noise reduction and edge enhancement. This algorithm could offer an important tool in enhancing DE image quality and potentially improving diagnostic performance.

Plate scatter correction for improved performance in dual-energy imaging.

We have developed a method for correcting for the scatter produced by the computed radiography plates in single-shot dual-energy imaging. In our technique, which uses a series of four computed radiography plates (the middle two plates act as a filter), the back plate can have scatter fractions as high as 30% due to the scatter from the first three plates. We use the deconvolution method to correct for this scatter, as the scatter medium is uniform and isotropic with a constant air gap. The ratio of the modulation transfer function (MTF) of a standard computed radiography plate to that of a dual-energy plate gives us the correction factor we need to scale the Fourier transforms of our images, thus compensating for the reduced response of the system due to scatter. We also investigated the effect of having an air gap between the back dual-energy plate and the first three plates. Our measurements show that this degrades the signal-to-noise response of the back plate slightly at low spatial frequencies. We also used independent scatter fraction measurements to verify the validity of our MTF measurements. Applying the correction to a back plate image improves the fine detail contrast by approximately 30%.