Search

PURPOSE Metabolic magnetic resonance imaging (MRI) using hyperpolarized [1-13 C]-pyruvate offers unprecedented new insight into disease and response to therapy. 13 C-enriched reference standards are required to enable fast and accurate calibration for 13 C studies, but care must be taken to ensure that the reference is compatible with both 13 C and 1 H acquisitions. The goal of this study was to optimize the composition of a 13 C-urea reference for a dual-tuned 13 C/1 H endorectal coil and minimize imaging artifacts in metabolic and multiparametric MRI studies involving hyperpolarized [1-13 C]-pyruvate. METHODS Due to a high amount of Gd doping for the purpose of reducing the spin-lattice relaxation time (T1 ) of urea, the 1 H signal produced by a reference of 13 C-urea in normal water was rapidly relaxed, resulting in severe artifacts in heavily T1 -weighted images. Hyperintense ringing artifacts in 1 H images were mitigated by reducing the 1 H concentration in a 13 C-urea reference via deuteration and lyophilization. Several references were fabricated and their SNR was compared using 1 H and 13 C imaging sequences on a 3T MRI scanner. Finally, 1 H prostate phantom imaging was conducted to compare image quality and 1 H signal intensity of normal and deuterated urea references. RESULTS The deuterated 13 C-urea reference provides strong 13 C signal for calibration and an attenuated 1 H signal that does not interfere with heavily T1 -weighted scans. Deuteration and lyophilization were fundamental to the reduction in 1 H signal and hyperintense ringing artifacts. There was a 25-fold reduction in signal intensity when comparing the nondeuterated reference to the deuterated reference, while the 13 C signal was unaffected. CONCLUSION A deuterated reference reduced hyperintense ringing artifacts in 1 H images by reducing the 1 H signal produced from the 13 C-urea in the reference. The deuterated reference can be used to improve anatomical image quality in future clinical 1 H and hyperpolarized [1-13 C]-pyruvate MRI prostate imaging studies.

Purpose: Studies have shown that Monte Carlo-based reconstruction could effectively improve the image quality of positron emission tomography. The authors have previously used a Gaussian rotator-based algorithm to efficiently reduce the computational cost for system matrix (SM) calculation and to meet the large memory requirements for SM storage. However, pronounced ringing artifacts were observed in the reconstructed image. In this article, the authors investigated and characterized these artifacts. Methods: The authors proposed an “ideal” rotator and used it as a baseline in the artifacts evaluation. This ideal rotator produces perfectly rotated images. The Gaussian rotator method was evaluated by a full systemmodel and a partial systemmodel without positron range and acolinearity, which could be compensated for by the blurring of the Gaussian rotator for F 18 studies. Noiseless data, Monte Carlo simulation data, as well as acquired experimental data were used to quantitatively characterize the behavior of the artifacts. Results: The study of the noiseless data indicated that the artifacts were mainly attributed to the rotator, which further blurred the simulated system responses. The simulation study suggested that the artifacts become less pronounced and not quantitatively significant in practice. This result is consistent with the experimental data study. Better contrast recovery was achieved with an overcompensated systemmodel. Traditionally, an undercompensated systemmodel was preferred to avoid artifacts. The authors’ studies suggest that the Gaussian rotator with the full systemmodel yields the best image quality among the evaluated methods with considerably reduced quantitative error and quantitatively insignificant artifacts in practice. Conclusions: The authors’ investigation indicated that a moderately overcompensated systemmodel (about 2 mm FWHM in this study) yielded better contrast recovery and quantitatively insignificant artifacts in practice.

Purpose: The purpose of this study is first to show the extent by which BLADE sequences can reduce all the image artifacts and second to verify that the usefulness of this technique in certain pathological conditions is significant. Methods: In this study, fourteen consecutive patients (5 females, 9 males), who routinely underwent MRI brain examination, between 2010–2014, were selected. The applied routine protocols for brain MR examination included the following sequences: 1) T2-W FLAIR axial; 2) T2-W TSE axial; 3) T2*-W axial, 4) T1-W TSE sagittal; 5) DWI-W axial; 6) T1-W TSE axial; 7) T1-W TSE axial+contrast. In cases of cystic tumors, the T2-W FLAIR BLADE sequence was added to the protocol. All the images were evaluated independently at two separate settings with 3 weeks interval by two radiologists. The radiologists also evaluated the presence of image artifacts (motion, flow, chemical shift, Gibbs ringing). To evaluate the size of the cyst, the two radiologists compared the two techniques (conventional and BLADE) by assessing the extent of the divergence in the measurements of the cysts. Results: Regarding the extent of the cyst size, BLADE measurements were found to be more reliable than the conventional ones with the differences being statistically significantmore » (p

Purpose: Grid artifacts are caused when using the antiscatter grid in obtaining digital x-ray images. In this paper, research on grid artifact reduction techniques is conducted especially for the direct detectors, which are based on amorphous selenium. Methods: In order to analyze and reduce the grid artifacts, the authors consider a multiplicative grid image model and propose a homomorphic filtering technique. For minimal damage due to filters, which are used to suppress the grid artifacts, rotated grids with respect to the sampling direction are employed, and min-max optimization problems for searching optimal grid frequencies and angles for given sampling frequencies are established. The authors then propose algorithms for the grid artifact reduction based on the band-stop filters as well as low-pass filters. Results: The proposed algorithms are experimentally tested for digital x-ray images, which are obtained from direct detectors with the rotated grids, and are compared with other algorithms. It is shown that the proposed algorithms can successfully reduce the grid artifacts for direct detectors. Conclusions: By employing the homomorphic filtering technique, the authors can considerably suppress the strong grid artifacts with relatively narrow-bandwidth filters compared to the normal filtering case. Using rotated grids also significantly reduces the ringing artifact. Furthermore, for specific grid frequencies and angles, the authors can use simple homomorphic low-pass filters in the spatial domain, and thus alleviate the grid artifacts with very low implementation complexity.

Purpose: TomoTherapy® treatment systems have exit MV detectors that record the radiation fluence during treatment. This information can be used in real time to verify delivereddose.Method and Materials: We developed a method that can extract the MLC leaf open time from the detector data without relying on the patient CT or pre‐measured database proposed by previous researchers. Because each leaf has penumbra, an open leaf will contribute signals to the detectors that correspond to adjacent leaves. Therefore, a closed leaf could be reported as open if its adjacent leaf is open. To eliminate the impact of the penumbra, we model the penumbra with a point‐spread‐function and apply an iterative deconvolution procedure to remove the penumbra. The iterative method successfully removes the penumbra without introducing ringing artifacts. The MLC leaf open time is then extracted with a simple thresholding method. To verify the algorithm, we install optical sensor on each MLC leaf that reports the leaf position. The leaf open time extracted from exit detector is compared to the optical sensor report. Results: The leaf open time extracted from the exit detector showed good agreement with the optical sensor under a variety of conditions, such as different attenuation phantom and different delivery sinogram. The detector measured leaf open time agrees with the optical sensor within 0.2 ms and a standard deviation of 3.6 ms. The impact of these uncertainties to the dose calculation is estimated to be around 0.2%. The algorithm is optimized so that it can run on the onboard computer in the machine to process the detector data in real time. Conclusion: The exit detectors in TomoTherapy® treatment systems can provide us information of leaf behavior during delivery. Together with the monitor chambers of the linac, the actual delivereddose to patient can be calculated.

Purpose: This paper discusses two methods for contrast enhancement of digital mammograms. One method is based on the Gaussian band‐pass filter and the other is based on the dyadic wavelet transform and multiscale non‐linear enhancement. Method and Materials: Band‐pass filter: A logarithmic tone mapping operator with adaptive base is applied to an image to elevate the lower intensities which lie mainly in the skin‐line zone. The mammogram is then filtered using a Gaussian band‐pass filter to enhance the structure of breast while suppressing noise and the slowly varying high density structure. Dyadic wavelet enhancement: Dyadic wavelet transform is applied to an image to decompose the image into horizontal detail and vertical detail subband images at different scales. Then, a non‐linear weighting function is used to modify the wavelet coefficients of each subband image followed by synthesis to yield the final enhanced image. The final enhanced image is obtained by synthesis of the modified subband images.Results: We compared the enhancement performance between these two approaches. Band‐pass filter does a better job of elevating the skin‐line zone to the intensity level of the interior of the breast and also enhancing the contrast. However, there are some ringing artifacts which lead to saturation in a few regions near the top and the bottom part of the image. The local contrast can be improved to emphasize both salient and subtle features of mammograms. However, the contrast of the skin‐line zone can not be enhanced well by the dyadic wavelet enhancement. Conclusion: Band‐pass filter is fast and can obtain a good result on skin‐line zone except for some ringing artifacts. Dyadic wavelet transform can enhance the unseen and barely seen features but is weak on skin‐line. We tested the above methods on MIAS and Fischer Imagingdatabases.

Purpose: To investigate the effect of different tone mapping operators in contrast enhancement of mammograms. A concave piecewise linear operator is proposed and is compared to a convex (logarithmic) and linear operator. Method and Materials: Tone mapping is performed prior to contrast enhancement to elevate the lower intensities that mainly correspond to the skin‐line zone. However, this leads to a loss in contrast due to compression of the gray levels. Gaussian band‐pass filtering is then performed to enhance the contrast of the mammogram. Filtering highlights the structure of breast while suppressing noise and the slowly varying high‐density structure. The filtering is done in the frequency domain and the image is mirrored and tiled prior to taking its Fourier transform to minimize edge effects. The tone mapping operators studied are logarithmic, linear and piecewise linear. The logarithmic operator has an adaptive base, the linear operator has constant slope in the intensities corresponding to the skin‐line zone, and the piecewise linear operator has varying slope in those intensities. Results: The enhanced images were viewed on a calibrated Dome C5i display. The logarithmic operator compresses the intensities near the threshold excessively which results in substantial loss of information. The linear and piecewise linear operators do not compress to the same extent. The piecewise linear operator compresses more in the range where there is not much information than in the range where there is more information. Also, tone mapping decreases the magnitude of the intensity step‐change from the breast region to the background. This reduces the ringing artifacts that result from the filtering operation Conclusion: The tone mapping operation improved the visual characteristics of the mammograms. Piecewise linear operator performed the best. The linear operator is comparable to the piecewise linear operator in most cases, though not as consistent. The logarithmic operator fared poorly.