Dixon techniques for water and fat imaging

To develop and evaluate a free-breathing chemical-shift-encoded (CSE) spoiled gradient-recalled echo (SPGR) technique for whole-heart water-fat imaging at 3 Tesla (T). We developed a three-dimensional (3D) multi-echo SPGR pulse sequence with electrocardiographic gating and navigator echoes and evaluated its performance at 3T in healthy volunteers (N = 6) and patients (N = 20). CSE-SPGR, 3D SPGR, and 3D balanced-SSFP with chemical fat saturation were compared in six healthy subjects with images evaluated for overall image quality, level of residual artifacts, and quality of fat suppression. A similar scoring system was used for the patient datasets. Images of diagnostic quality were acquired in all but one subject. CSE-SPGR performed similarly to SPGR with fat saturation, although it provided a more uniform fat suppression over the whole field of view. Balanced-SSFP performed worse than SPGR-based methods. In patients, CSE-SPGR produced excellent fat suppression near metal. Overall image quality was either good (7/20) or excellent (12/20) in all but one patient. There were significant artifacts in 5/20 clinical cases. CSE-SPGR is a promising technique for whole-heart water-fat imaging during free-breathing. The robust fat suppression in the water-only image could improve assessment of complex morphology at 3T and in the presence of off-resonance, with additional information contained in the fat-only image.

To develop a robust 3D fast spin echo (FSE) T(2)-weighted imaging method with uniform water and fat separation in a single acquisition, amenable to high-quality multiplanar reformations. The Iterative Decomposition of water and fat with Echo Asymmetry and Least squares estimation (IDEAL) method was integrated with modulated refocusing flip angle 3D-FSE. Echoes required for IDEAL processing were acquired by shifting the readout gradient with respect to the Carr-Purcell-Meiboom-Gill echo. To reduce the scan time, an alternative data acquisition using two gradient echoes per repetition was implemented. Using the latter approach, a total of four gradient echoes were acquired in two repetitions and used in the modified IDEAL reconstruction. 3D-FSE T(2)-weighted images with uniform water-fat separation were successfully acquired in various anatomies including breast, abdomen, knee, and ankle in clinically feasible scan times, ranging from 5:30-8:30 minutes. Using water-only and fat-only images, in-phase and out-of-phase images were reconstructed. 3D-FSE-IDEAL provides volumetric T(2)-weighted images with uniform water and fat separation in a single acquisition. High-resolution images with multiple contrasts can be reformatted to any orientation from a single acquisition. This could potentially replace 2D-FSE acquisitions with and without fat suppression and in multiple planes, thus improving overall imaging efficiency.

Previously published fast spin-echo (FSE) implementations of a Dixon method for water and fat separation all require multiple scans and thus a relatively long scan time. Further, the minimum echo spacing (esp), a time critical for FSE image quality and scan efficiency, often needs to be increased in order to bring about the required phase shift between the water and fat signals. This work proposes and implements a novel FSE triple-echo Dixon (fTED) technique that can address these limitations. In the new technique, three raw images are acquired in a single FSE scan by replacing each frequency-encoding gradient in a conventional FSE with three consecutive gradients of alternating polarity. The timing of the three gradients is adjusted by selecting an appropriate receiver bandwidth (RBW) so that the water and fat signals for the three corresponding echoes have a relative phase shift of -180 degrees , 0 degrees , and 180 degrees , respectively. A fully automated postprocessing algorithm is then used to generate separate water-only and fat-only images for each slice. The technique was implemented with and without parallel imaging. We demonstrate that the new fTED technique enables both uniform water/fat separation and fast scanning with uncompromised scan parameters, including applications such as T(2)-weighted separate water and fat imaging of the abdomen during breath-holding.

The purposes of this study were to determine optimum flip angles (FAs) and to compare the effectiveness of fat suppression and signal homogeneity among three techniques, spectral attenuated with inversion recovery (SPAIR), principle of selective excitation technique (PROSET), and three-point DIXON technique (DIXON), of the bilateral breast dynamic sequence acquired using the optimum FA at 3.0 T. Using a homemade phantom that represented a tumor, fat, and a mammary gland, the optimum FAs were determined from the change of fat signal intensity, signal-to-noise ratio (SNR) of the mammary gland, and contrast ratio (CR) between the tumor and mammary gland. The effectiveness of fat suppression and signal homogeneity were compared in ten breast cancer cases, using the CR between fat and pectoralis muscle signal intensities and the standard deviation (SD) of fat signal intensity, respectively. The optimum FAs for SPAIR, PROSET, and DIXON were 10, 20, and 20 degrees, respectively. The mean CR between fat and pectoralis muscle signal intensities achieved using SPAIR, PROSET, and DIXON were 0.19, 0.30 and 0.40, respectively, and the mean SDs of the fat signal intensities were 90.2, 103.1, and 30.5, respectively. The DIXON technique provided better fat suppression and signal homogeneity than the other two techniques. The results of this study suggest the possible application of the DIXON technique in combination with the optimum FA setting as an effective fat suppression technique for the bilateral breast dynamic sequence at 3.0 T.

Localized Larmor frequency-guided fat and water proton MRI of the spine: A method to emphasize pathological findings

PurposeTo develop a fat‐water imaging method that allows reliable separation of the two tissues, uses established robust reconstruction methods, and requires only one single‐echo acquisition.Theory and MethodsThe proposed method uses spectrally selective dual‐band excitation in combination with CAIPIRINHA to generate separate images of fat and water simultaneously. Spatially selective excitation without cross‐contamination is made possible by the use of spatial‐spectral pulses. Fat and water images can either be visualized separately, or the fat images can be corrected for chemical shift displacement and, in gradient echo imaging, for chemical shift‐related phase discrepancy, and recombined with water images, generating fat‐water images free of chemical shift effects. Gradient echo and turbo spin echo sequences were developed based on this Simultaneous Multiple Resonance Frequency imaging (SMURF) approach and their performance was assessed at 3Tesla in imaging of the knee, breasts, and abdomen.ResultsThe proposed method generated well‐separated fat and water images with minimal unaliasing artefacts or cross‐excitation, evidenced by the near absence of water signal attributed to the fat image and vice versa. The separation achieved was similar to or better than that using separate acquisitions with water‐ and fat‐saturation or Dixon methods. The recombined fat‐water images provided similar image contrast to conventional images, but the chemical shift effects were eliminated.ConclusionSimultaneous Multiple Resonance Frequency imaging is a robust fat‐water imaging technique that offers a solution to imaging of body regions with significant amounts of fat.

The Spiral two-point Dixon (Spiral 2PD) technique has recently been proposed as a method for unambiguous water-fat decomposition in spiral imaging. It also corrects for off-resonance blurring artifacts using only two data sets. In the Spiral 2PD technique, several predetermined off-resonance frequencies are tested to both separate water and fat signals and deblur the decomposed images. Unfortunately, the algorithm is computationally quite intensive since the range of tested frequencies must be set sufficiently large to span the full range of anticipated B(0) variation over the scanned objects. The block regional off-resonance correction (BRORC) algorithm corrects for off-resonance blurring artifacts block by block through the reconstructed image and usually provides several times higher computational efficiency than the conventional frequency-segmented off-resonance correction algorithm. This work shows that both water-fat decomposition and blurring artifact correction can be performed block by block using two spiral images with different TEs and that this new technique (BRORC-Spiral2PD technique) significantly improves the computational efficiency of other Spiral 2PD algorithms, opening new opportunities for spiral imaging.

In order to satisfy the Carr-Purcell-Meiboom-Gill (CPMG) condition, echo shift as dictated in fast-spin-echo (FSE)-based Dixon imaging was previously achieved by applying a time shift to the readout gradient and the data acquisition window. Accordingly, interecho spacing is increased, which entails increased image blurring and, in multislice imaging, a significant reduction in the slice coverage for a given imaging time. In this work, a new method is developed by which the echo shift is induced by "sandwiching" in time the readout gradient with a pair of small gradients of equal area and of opposite polarity. While data with non-zero phase shifts between water and fat signals are collected as fractional echoes, no increase in echo spacing is necessary with the modified acquisition strategy, and increased time efficiency is therefore achieved. In order to generate separate water-only and fat-only images in data processing, a set of low-resolution images are first reconstructed from the central symmetric portion (either 128 x 128 or 64 x 64) of the acquired multipoint Dixon data. High-resolution images using all the acquired data, including some partial Fourier-reconstructed images, are then phase demodulated using the phase errors determined from the low-resolution images. The feasibility of the technique is demonstrated using a water and fat phantom as well as in clinical patient imaging.

Large and spatially-linear phase errors along the frequency-encode direction may be induced by several common and hard-to-avoid system imperfections such as eddy currents. For data acquired in dual-echo Dixon techniques, the linear phase error can be more aggravated when compared to that acquired in a single echo and can pose challenges to a phase-correction algorithm necessary for successful Dixon processing. In this work, we propose a two-step process that first corrects the linear component of the phase errors with a modified Ahn-Cho algorithm (Ahn CB and Cho ZH, IEEE Trans. Med. Imaging 6:32, 1987) and then corrects the residual phase errors with a previously-developed region-growing algorithm (Ma J, Magn. Res. Med. 52:415, 2004). We demonstrate that successive application of the two-step process to data from a dual-echo Dixon technique provides a "1-2 punch" to the overall phase errors and can overcome local water and fat separation failures that are observed when the region-growing-based algorithm is applied alone.

Signal from fat is normally removed from MR images either by fat separation techniques that distinguish water from fat signal after the data has been received, or by fat suppression techniques that prevent the fat signal from being received. Most approaches to fat separation are variations on Dixon imaging. The primary downside to Dixon imaging is the requirement for multiple images with stationary anatomy, often with specific TEs. An alternate approach is to take only one image, estimate phase errors to correct for inhomogeneity or other effects, and then separate the water and fat using the known phase shift. This has shown promise in previously published work, but the water and fat signals were always perpendicular, requiring a fixed TE. We consider the possibility of separation from a single, phase-corrected image with an arbitrary angle between water and fat signals. We note that a change of basis will separate water and fat signals into two images with additive zero-mean Gaussian noise. However, as the angle between water and fat nears pi or 0, the noise power in the separated images increases rapidly. We discuss techniques for reducing this noise magnification.

To evaluate a prototype fast spin-echo (FSE) triple-echo Dixon (FTED) technique for T2-weighted spine imaging with and without fat suppression compared to conventional T2-weighted fast recovery (FR) FSE and short-tau inversion recovery (STIR) imaging. Sixty-one patients were referred for spine magnetic resonance imaging (MRI) including sagittal FTED (time 2:26), STIR (time 2:42), and T2 FRFSE (time 2:55). Two observers compared STIR and FTED water images and T2 FRFSE and FTED T2 images for overall image quality, fat suppression, anatomic sharpness, motion, cerebrospinal fluid (CSF) flow artifact, susceptibility, and disease depiction. On FTED images water and fat separation was perfect in 58 (.95) patients. Compared to STIR, the FTED water images demonstrated less motion in 57 (.93) of 61 patients (P < 0.05), better anatomic sharpness in 51 (.84) and patients (P < 0.05), and less CSF flow artifact in 7 (.11) P < 0.05) patients. There was no difference in fat suppression or chemical shift artifact. T2 FRFSE and FTED T2 images showed equivalent motion, CSF flow, and chemical shift artifact. Lesion depiction was equivalent on FTED water and STIR images and FTED T2 and T2 FRFSE images. FTED efficiently provides both fat-suppressed and nonfat-suppressed T2-weighted spine images with excellent image quality, equal disease depiction, and 56% reduction in scan time compared to conventional STIR and T2 FRFSE.

Efficient Water-Fat Separation for Extremity MRI at Ultra-Low-Field (0.05T).

A subjective and objective comparison of tissue contrast and imaging artifacts present in routine spin echoes and in iterative decomposition of asymmetric spin echoes for soft tissue neck MRI

RAD-GRASE is an MRI sequence that combines radial (RAD) k-space scanning with the gradient and spin-echo (GRASE) technique. RAD-GRASE has the advantages of all radial data acquisition methods in that it can reduce motion sensitivity and correct motion-induced data errors, which can be exploited to achieve high-resolution diffusion-weighted imaging (DWI). One can obtain different types of image contrast, including DWI, T(1), T(2), and T(2)*, in RAD-GRASE by controlling the magnetization preparation and sequence timing. Moreover, because there is oversampling of the low spatial frequencies inherent to radial sequences, partial data reconstruction can be used to achieve multiple forms of image contrast from a single acquired data set, and to generate parametric image maps of equilibrium magnetization, T(2), and T(2) (dagger). The RAD-GRASE technique can also be used to achieve fat-suppressed and/or separated fat and water images by choosing the appropriate timing parameters.

To evaluate the possibility of quantifying brown adipose tissue (BAT) volume and fat concentration with a high resolution, long echo time, dual-echo Dixon imaging protocol. A 0.42 mm isotropic resolution water-fat separated MRI protocol was implemented by using the second opposite-phase echo and third in-phase echo. Fat images were calibrated with regard to the intensity of nearby white adipose tissue (WAT) to form relative fat content (RFC) images. To evaluate the ability to measure BAT volume and RFC contrast dynamics, rats were divided into two groups that were kept at 4° or 22°C for 5 days. The rats were then scanned in a 70 cm bore 3.0 Tesla MRI scanner and a human dual energy CT. Interscapular, paraaortal, and perirenal BAT (i/pa/pr-BAT) depots as well as WAT and muscle were segmented in the MRI and CT images. Biopsies were collected from the identified BAT depots. The biopsies confirmed that the three depots identified with the RFC images consisted of BAT. There was a significant linear correlation (P < 0.001) between the measured RFC and the Hounsfield units from DECT. Significantly lower iBAT RFC (P = 0.0064) and significantly larger iBAT and prBAT volumes (P = 0.0017) were observed in the cold stimulated rats. The calibrated Dixon images with RFC scaling can depict BAT and be used to measure differences in volume, and fat concentration, induced by cold stimulation. The high correlation between RFC and HU suggests that the fat concentration is the main RFC image contrast mechanism.