Conventional GRAPPA Research Articles

Fuzzy ripple artifact in high resolution fMRI: identification, cause, and mitigation.

High resolution fMRI is a rapidly growing research field focused on capturing functional signal changes across cortical layers. However, the data acquisition is limited by low spatial frequency EPI artifacts; termed here as Fuzzy Ripples. These artifacts limit the practical applicability of acquisition protocols with higher spatial resolution, faster acquisition speed, and they challenge imaging in lower brain areas. We characterize Fuzzy Ripple artifacts across commonly used sequences and distinguish them from conventional EPI Nyquist ghosts, off-resonance effects, and GRAPPA artifacts. To investigate their origin, we employ dual polarity readouts. Our findings indicate that Fuzzy Ripples are primarily caused by readout-specific imperfections in k-space trajectories, which can be exacerbated by inductive coupling between third-order shims and readout gradients. We also find that these artifacts can be mitigated through complex-valued averaging of dual polarity EPI or by disconnecting the third-order shim coils. The proposed mitigation strategies allow overcoming current limitations in layer-fMRI protocols: (1)Achieving resolutions beyond 0.8mm is feasible, and even at 3T, we achieved 0.53mm voxel functional connectivity mapping.(2)Sub-millimeter sampling acceleration can be increased to allow sub-second TRs and laminar whole brain protocols with up to GRAPPA 8.(3)Sub-millimeter fMRI is achievable in lower brain areas, including the cerebellum.

Echo planar time-resolved imaging with subspace reconstruction and optimized spatiotemporal encoding.

To develop new encoding and reconstruction techniques for fast multi-contrast/quantitative imaging. The recently proposed Echo Planar Time-resolved Imaging (EPTI) technique can achieve fast distortion- and blurring-free multi-contrast/quantitative imaging. In this work, a subspace reconstruction framework is developed to improve the reconstruction accuracy of EPTI at high encoding accelerations. The number of unknowns in the reconstruction is significantly reduced by modeling the temporal signal evolutions using low-rank subspace. As part of the proposed reconstruction approach, a B0 -update algorithm and a shot-to-shot B0 variation correction method are developed to enable the reconstruction of high-resolution tissue phase images and to mitigate artifacts from shot-to-shot phase variations. Moreover, the EPTI concept is extended to 3D k-space for 3D GE-EPTI, where a new "temporal-variant" of CAIPI encoding is proposed to further improve performance. The effectiveness of the proposed subspace reconstruction was demonstrated first in 2D GESE EPTI, where the reconstruction achieved higher accuracy when compared to conventional B0 -informed GRAPPA. For 3D GE-EPTI, a retrospective undersampling experiment demonstrates that the new temporal-variant CAIPI encoding can achieve up to 72× acceleration with close to 2× reduction in reconstruction error when compared to conventional spatiotemporal-CAIPI encoding. In a prospective undersampling experiment, high-quality whole-brain and tissue phase maps at 1 mm isotropic resolution were acquired in 52 seconds at 3T using 3D GE-EPTI with temporal-variant CAIPI encoding. The proposed subspace reconstruction and optimized temporal-variant CAIPI encoding can further improve the performance of EPTI for fast quantitative mapping.

Self-calibrated interpolation of non-Cartesian data with GRAPPA in parallel imaging.

To develop a non-Cartesian k-space reconstruction method using self-calibrated region-specific interpolation kernels for highly accelerated acquisitions. In conventional non-Cartesian GRAPPA with through-time GRAPPA (TT-GRAPPA), the use of region-specific interpolation kernels has demonstrated improved reconstruction quality in dynamic imaging for highly accelerated acquisitions. However, TT-GRAPPA requires the acquisition of a large number of separate calibration scans. To reduce the overall imaging time, we propose Self-calibrated Interpolation of Non-Cartesian data with GRAPPA (SING) to self-calibrate region-specific interpolation kernels from dynamic undersampled measurements. The SING method synthesizes calibration data to adapt to the distinct shape of each region-specific interpolation kernel geometry, and uses a novel local k-space regularization through an extension of TT-GRAPPA. This calibration approach is used to reconstruct non-Cartesian images at high acceleration rates while mitigating noise amplification. The reconstruction quality of SING is compared with conjugate-gradient SENSE and TT-GRAPPA in numerical phantoms and in vivo cine data sets. In both numerical phantom and in vivo cine data sets, SING offers visually and quantitatively similar reconstruction quality to TT-GRAPPA, and provides improved reconstruction quality over conjugate-gradient SENSE. Furthermore, temporal fidelity in SING and TT-GRAPPA is similar for the same acceleration rates. G-factor evaluation over the heart shows that SING and TT-GRAPPA provide similar noise amplification at moderate and high rates. The proposed SING reconstruction enables significant improvement of acquisition efficiency for calibration data, while matching the reconstruction performance of TT-GRAPPA.

Accelerated imaging with segmented 2D pulses using parallel imaging and virtual coils.

Large magnetic field inhomogeneity can be a significant cause of spatial flip-angle variation when using ordinary, limited-bandwidth RF pulses. Multidimensional RF pulses are particularly sensitive to inhomogeneity due to their extended pulse length, which decreases their bandwidth. Previously, it was shown that, by breaking a 2D pulse into multiple undersampled k-space segments, the excitation bandwidth can be increased at the expense of increased imaging time. The present study shows how this increased imaging time can be offset by undersampling acquisition k-space in a phase-encoded dimension that is in the direction of excitation segmentation. Data from each segment are viewed as originating from "virtual receive coils" rather than multiple physical coils. The undersampled data are reconstructed using parallel imaging techniques (e.g. as in GRAPPA). The method was tested in vivo with brain imaging at both 3 T and 4 T, and used in conjunction with a 32-channel head coil and conventional GRAPPA on the 3 T data. Relationships with existing techniques and future applications are discussed.

Improving GRAPPA reconstruction using joint nonlinear kernel mapped and phase conjugated virtual coils

To improve the reconstruction condition and alleviate the noise amplification of GRAPPA reconstruction by aggregating the phase conjugated and nonlinear kernel mapped coils with the original physical coil.Nonlinear GRAPPA (NL-GRAPPA) is a kernel-based non-iterative approach which can reduce noise-induced error in GRAPPA reconstruction. And virtual conjugate coil (VCC) embeds the conjugate symmetric property of k-space into GRAPPA data synthesis to improve reconstruction condition. This work proposed NL-VCC-GRAPPA to jointly utilize the nonlinear mapped virtual coil and phase conjugated virtual coil to further reduce noise amplification in parallel imaging. In vivo static and dynamic 2D imaging accelerated by uniform undersampling schemes were performed to evaluate the proposed method in terms of visual image quality, root-mean-square-error (RMSE), and geometry factor (g-factor). The effects of acceleration factors, calibration data size and kernel shape on the proposed model were also separately analyzed and discussed.The proposed method illustrated improved visual image quality evidenced by reduced retrospective RMSE and prospective g-factor comparing with conventional GRAPPA and the recently proposed iterative SENSE-LORAKS reconstructions. Although a larger amount of calibration data and smaller kernel size were required to stabilize the calibration of fourfold extended kernel for the proposed method, it was non-iterative and relatively insensitive to parameter adjustment in the applications.The proposed NL-VCC-extension to conventional GRAPPA brings visible improvements for imaging scenarios accelerated by the widely available uniform undersampling schemes in a practically efficient manner without iteration.

Highly-accelerated volumetric brain examination using optimized wave-CAIPI encoding.

Rapid volumetric imaging protocols could better utilize limited scanner resources. To develop and validate an optimized 6-minute high-resolution volumetric brain MRI examination using Wave-CAIPI encoding. Prospective. Ten healthy subjects and 20 patients with a variety of intracranial pathologies. At 3 T, MPRAGE, T2 -weighted SPACE, SPACE FLAIR, and SWI were acquired at 9-fold acceleration using Wave-CAIPI and for comparison at 2-4-fold acceleration using conventional GRAPPA. Extensive simulations were performed to optimize the Wave-CAIPI protocol and minimize both g-factor noise amplification and potential T1 /T2 blurring artifacts. Moreover, refinements in the autocalibrated reconstruction of Wave-CAIPI were developed to ensure high-quality reconstructions in the presence of gradient imperfections. In a randomized and blinded fashion, three neuroradiologists assessed the diagnostic quality of the optimized 6-minute Wave-CAIPI exam and compared it to the roughly 3× slower GRAPPA accelerated protocol using both an individual and head-to-head analysis. A noninferiority test was used to test whether the diagnostic quality of Wave-CAIPI was noninferior to the GRAPPA acquisition, with a 15% noninferiority margin. Among all sequences, Wave-CAIPI achieved negligible g-factor noise amplification (gavg ≤ 1.04) and burring artifacts from T1 /T2 relaxation. Improvements of our autocalibration approach for gradient imperfections enabled increased robustness to gradient mixing imperfections in tilted-field of view (FOV) prescriptions as well as variations in gradient and analog-to-digital converter (ADC) sampling rates. In the clinical evaluation, Wave-CAIPI achieved similar mean scores when compared with GRAPPA (MPRAGE: ØW = 4.03, ØG = 3.97; T2 w SPACE: ØW = 4.00, ØG = 4.00; SPACE FLAIR: ØW = 3.97, ØG = 3.97; SWI: ØW = 3.93, ØG = 3.83) and was statistically noninferior (N = 30, P < 0.05 for all sequences). The proposed volumetric brain exam retained comparable image quality when compared with the much longer conventional protocol. 2 Technical Efficacy: Stage 1 J. Magn. Reson. Imaging 2019;50:961-974.

MultiNet PyGRAPPA: Multiple neural networks for reconstructing variable density GRAPPA (a 1H FID MRSI study)

Magnetic resonance spectroscopic imaging (MRSI) is a powerful tool for mapping metabolite levels across the brain, however, it generally suffers from long scan times. This severely hinders its application in clinical settings. Additionally, the presence of nuisance signals (e.g. the subcutaneous lipid signals close to the skull region in brain metabolite mapping) makes it challenging to apply conventional acceleration techniques to shorten the scan times. The goal of this work is, therefore, to increase the overall applicability of high resolution metabolite mapping using 1H MRSI by introducing a novel GRAPPA acceleration acquisition/reconstruction technique. An improved reconstruction method (MultiNet) is introduced that uses machine learning, specifically neural networks, to reconstruct accelerated data. The method is further modified to use more neural networks with nonlinear hidden layers and is then combined with a variable density undersampling scheme (MultiNet PyGRAPPA) to enable higher in-plane acceleration factors of R = 5.6 and R = 7 for a non-lipid suppressed ultra-short TR and TE 1H FID MRSI sequence. The proposed method is evaluated for high resolution metabolite mapping of the human brain at 9.4T. The results show that the proposed method is superior to conventional GRAPPA: there is no significant residual lipid aliasing artifact in the images when the proposed MultiNet method is used. Furthermore, the MultiNet PyGRAPPA acquisition/reconstruction method with R = 5.6 results in reproducible high resolution metabolite maps (with an in-plane matrix size of 64 × 64) that can be acquired in 2.8 min on 9.4T. In conclusion, using multiple neural networks to predict the missing points in GRAPPA reconstruction results in a more reliable data recovery while keeping the noise levels under control. Combining this high fidelity reconstruction with variable density undersampling (MultiNet PyGRAPPA) enables higher in-plane acceleration factors even for non-lipid suppressed 1H FID MRSI, without introducing any structured aliasing artifact in the image.

Accelerated 3D-GRASE imaging improves quantitative multiple post labeling delay arterial spin labeling.

To investigate the impact of accelerated, single-shot 3D-GRASE acquisition on quantitative arterial spin labeling (ASL) with multiple and single post-labeling delay (PLD) in terms of perfusion-weighted SNR per unit scan time (TSNRPW ) and quantification accuracy. Five subjects were scanned on a 3T MRI scanner using the pseudo-continuous arterial spin labeling (PCASL) technique with a 3D-GRASE imaging sequence capable of parallel imaging acceleration. A 3-inversion pulse background suppression was simulated and implemented in the sequence. Three time-matched single PLD measurements, a segmented one without acceleration, 1 with conventional GRAPPA, and 1 with CAIPIRINHA sampling, were used to compare TSNRPW . Three time-matched multiple PLD measurements with the identical imaging parameters were additionally evaluated (no acceleration vs. CAIPIRINHA sampling vs. CAIPIRINHA sampling with doubled number of PLDs). Cerebral blood flow and arterial transit time fit uncertainties were compared and used as a quality measure. The single PLD measurements show an 11% TSNRPW increase using CAIPIRINHA sampling instead of GRAPPA sampling, while the non-accelerated scan exhibits 35% higher TSNRPW compared to the GRAPPA scan. However, taking advantage of the increased number of averages for multiple PLD acquisitions, a 14%/16% (gray matter) and 34%/36% (white matter) reduction of CBF fit uncertainty is observed with CAIPIRINHA sampling (6 PLDs/12 PLDs) compared to no acceleration. Accelerated single-shot 3D-GRASE with PCASL allows for smaller quantification uncertainties than time-matched segmented acquisitions. Corresponding single-shot acquisitions with acceptable blurring and no intra-volume motion render state-of-the-art ASL methods in a clinically feasible time possible.

Accelerated 3D bSSFP imaging for treatment planning on an MRI-guided radiotherapy system.

The purpose of this study was to introduce a compressed sensing and parallel imaging-combined technique to reduce the acquisition time of planning MRI for MR-guided radiotherapy (MRgRT) systems. A variable-density Poisson-Disk (VDPD) undersampling acquisition along with compressed sensing reconstruction technique was developed and compared with the current planning MR protocol, which uses an optimized balanced steady-state free precession sequence with 7.5-fold (7.5×) acceleration achieved by GRAPPA and partial Fourier. The image quality of GRAPPA and VDPD with 7.5× and 15× acceleration was compared with fully sampled images on a phantom. Two volunteers were recruited to compare the invivo imaging performance. Ten patients with abdominal tumors were scanned using the conventional GRAPPA 7.5× (25s) and the proposed VDPD 15× (12.5s) sequences. Three readers scored the two approaches in terms of the quality for organ and tumor delineation. The gross tumor volume (GTV) and two kidneys were contoured. Differences in centroid location and contour volumes, Dice coefficients, and mean distance-to-agreement (MDA) between contours draw on the two techniques were calculated. All studies were performed on a 0.35T MRgRT system. In the phantom study, VDPD with 15× acceleration rate had lower noise level than GRAPPA with 7.5× acceleration. In both the phantom and volunteer study, noise amplification was apparent when the acceleration rate was increased from 7.5× to 15× in the GRAPPA acquisition, whereas it was minimally increased using the VDPD approach. In the patient study, no significant difference was found for the scoring and contouring statistics between the two techniques, whereas VDPD only took half the scan time as GRAPPA. Volume difference for the GTV and two kidneys between GRAPPA 7.5× and VDPD 15× was around 7.6%, 1.3%, and 2.8%, respectively; while the Dice index was approximately 0.85, 0.92, and 0.90, respectively. The proposed technique reduced the acquisition time by half and provided comparable or improved image quality than the standard planning MRI protocol.

High resolution CBV assessment with PEAK-EPI: k-t-undersampling and reconstruction in echo planar imaging.

Achieving higher spatial resolution and improved brain coverage while mitigating in-plane susceptibility artifacts in the assessment of perfusion parameters, such as cerebral blood volume, in echo planar imaging (EPI)-based dynamic susceptibility contrast weighted cerebral perfusion measurements. PEAK-EPI, an EPI sequence with interleaved readout trajectories and three different strategies for autocalibration-signal acquisition (inplace, dynamic extra and extra) is presented. Performance of each approach is analyzed in vivo based on flip angle variation induced dynamics, assessing temporal fidelity, temporal SNR and g-factors. All approaches are compared with conventional GRAPPA reconstructions. PEAK-EPI with inplace autocalibration-signal at R = 5 is then compared with the standard clinical EPI protocol in six patients, using two half-dose dynamic susceptibility contrast weighted cerebral perfusion measurements per subject. PEAK-EPI acquisition facilitates a substantial increase of spatial resolution at a higher number of slices per TR and provides improved SNR compared to conventional GRAPPA. High dependency of the resulting reconstruction quality on the type of autocalibration-signal acquisition is observed. PEAK-EPI with inplace autocalibration-signal achieves high temporal fidelity and initial feasibility is shown. The obtained high resolution cerebral blood volume maps reveal more detailed information than in corresponding standard EPI measurements and facilitate detailed delineation of tumorous tissue. Magn Reson Med 77:2153-2166, 2017. © 2016 International Society for Magnetic Resonance in Medicine.

Penerapan Teknik Parallel Imaging Pada Pesawat MRI 0,35 Tesla Untuk Optimalisasi Kualitas Informasi Anatomi Pada MRI Lumbal Pembobotan T1WI dan T2WI Potongan Sagital

Background : Parallel imaging is one of the MRI Scanning techniques used to reduce the overall scan time when the patients with unvoluntary movement being examined with a low magnetic field of 0,35 T. This research aims to determine the difference between the clinical image quality of the conventional turbo spin echo (TSE) with mSENSE and that of the TSE with GRAPPA parallel imaging techniques from which resulting the MRI T1 and T2 Weighted Images (T1WI and T2WI) sagittal view of lumbar spines, and to define the techniques that clinically provide the most approriate anatomical information.Methods : This experimental study is made performed by the MRI 0.35 T in which 10 patients who had hernia nucleus pulposus (HNP) desease participated in the experiments ramdomly. The appointed Radiologists blended in the image evaluation using an image checklist to assess the visualisation of anatomical organs on the resulted sagittal lumbar MRI T1WI and T2WI. The two non-parametric statistical tools, Friedman test and the post hoc Wilcoxon matched pairs test, is used to analyze all the data descriptively. Testing the resesearch hypotheses with 95% of confident interval is to proved the differences between resulted sagittal lumbar MRI T1WI and T2WI..Results : The results shown there is a significant difference on the image quality of anatomical information when conventional TSE, parallel imaging-mSENSE and -GRAPPA, with T1WI are applied in the imaging techniques. When those imaging techniques are employed to obtain T2WI, the result is not significant in contrast.Conclusion : Good imaging techniques with adequate clinical image quality are ranked sequently as the conventional TSE, the mSENSE and GRAPPA.

Accelerated model-based proton resonance frequency shift temperature mapping using echo-based GRAPPA reconstruction

PurposeTo develop an acceleration method for MR temperature estimation using model-based proton resonance frequency (PRF) shift method. Materials and MethodsImages of 16 different echo times (TE) were acquired in one RF excitation using a multi-echo gradient-recalled echo (GRE) sequence. Fully sampled k-space data were retrospectively under-sampled at a net reduction factor between two and three using the proposed under-sampling strategy. K-spaces of three different TEs were combined together to perform the proposed reconstruction method called Echo-based GRAPPA. Ex vivo goose liver cooling experiment and in vivo breast imaging experiment were performed to investigate the accuracy of Echo-based GRAPPA. Conventional GRAPPA reconstruction was implemented for comparison using the same sampling pattern. ResultsThe goose liver imaging experiment shows that the reconstruction-induced temperature RMSE of a selected region of interest (ROI) is less than 1.4°C for Echo-based GRAPPA at a net reduction factor of 2.3. The breast imaging experiment shows that the mean temperature error of water–fat mixed ROIs is 2.3°C at a net reduction factor of 2.7. Conventional GRAPPA shows larger temperature RMSE than Echo-based GRAPPA. ConclusionThe proposed method can accelerate the MR temperature estimation using model-based PRF at a net reduction factor between two and three with a reconstruction-induced temperature error less than 3°C in water–fat mixed ROIs.

Biventricular strain analysis at 1.5T cardiac MR imaging: preliminary results in volunteers using an iterative SENSE reconstruction with L1 regularization

Background Changes in myocardial strain have been shown to precede onset of systolic dysfunction in patients with cardiomyopathy. Traditionally performed with echocardiography, acoustic windows can limit strain evaluation. Preliminary work has shown good agreement between myocardial strain derived from deformation field analysis at balanced steady-state free-precession (bSSFP) cine MRI and speckle tracking echocardiography. The application of a novel prototype iterative SENSE reconstruction with L1 regularization (IRSENSE) to highly accelerated segmented bSSFP cine acquisitions can maintain or improve the effective temporal resolution with a shorter breath-hold. The purpose of this study is to evaluate left- and right-ventricular stain analysis from bSSFP cine acquisitions using conventional GRAPPA and highly accelerated T-PAT with IR-SENSE. We hypothesize that myocardial strain parameters derived from accelerated T-PAT cine acquisitions with IR-SENSE are similar to those obtained using conventional segmented bSSFP with comparable effective temporal resolutions.

Clinical application of 3D VIBECAIPI-DIXON for non-enhanced imaging of the pancreas compared to a standard 2D fat-saturated FLASH

PurposeTo compare a fast 3D VIBE sequence with Dixon fat saturation and CAIPIRINHA acceleration techniques (3D VIBECAIPI-DIXON) to a standard 2D FLASH sequence with spectral fat saturation and conventional GRAPPA acceleration technique (2D FlashGRAPPA-fs) for non-enhanced imaging of the pancreas. Methods and materialsIn this retrospective, institutional review board-approved intra-individual comparison study, 29 patients (7 female, 22 male; mean age 60.4±20.9 years) examined on a 48-channel 3.0-T MR system (MAGNETOM Skyra VD 13, Siemens Healthcare Sector, Germany) were included. 3D VIBECAIPI-DIXON (TR/TE−3.95/2.5+1.27 ms; spatial resolution−1.2×1.2×3.0 mm3; CAIPIRINHA 2×2 [1], acquisition time−0:12 min) and 2D FlashGRAPPA-fs (TR/TE−195/3.69 ms; 1.2×1.2×3.0 mm3; GRAPPA 2, 3×0:21 min) sequences were performed in each subject in random order prior to the administration of an intravenous contrast agent. Two radiologists evaluated the images with regard to diagnostic preference. Semi-quantitative signal ratios were calculated for the pancreas versus the liver, spleen, muscle, and visceral fat. Inter-reader agreement was calculated using unweighted Cohen's kappa. Signal ratio results were analyzed using a univariate analysis of variance. Additional signal-to-noise (SNR) measurements were performed in a phantom. Results3D VIBECAIPI-DIXON was preferred in 72.4% (both readers) and 2D FlashGRAPPA-fs in 3.4%/6.9% (reader 1/2) of cases with a kappa value of 0.756. The main reasons for this preference were homogenous fat saturation with 3D VIBECAIPI-DIXON and reduced motion artifacts due to a faster acquisition, leading to improved delineation of the pancreas. Signal ratios of pancreatic to fat signal for 3D VIBECAIPI-DIXON (10.08±3.48) and 2D FlashGRAPPA-fs (6.53±3.07) were statistically different (P<.001). However, no additional statistically significant differences in signal ratios were identified (range: 0.73±0.18 to 1.37±0.40; .514<P<.961). SNR did not statistically significantly differ between the sequences. Conclusion3D VIBECAIPI-DIXON enables robust pancreatic imaging with a shorter time and improved fat suppression relative to conventional 2D FlashGRAPPA-fs. At an acquisition time of 12 seconds, 3D VIBECAIPI-DIXON can be obtained in considerably less time than standard fat-saturated VIBE sequences.

Increasing efficiency of parallel imaging for 2D multislice acquisitions

Parallel imaging algorithms require precise knowledge about the spatial sensitivity variation of the receiver coils to reconstruct images with full field of view (FOV) from undersampled Fourier encoded data. Sensitivity information must either be given a priori, or estimated from calibration data acquired along with the actual image data. In this study, two approaches are presented, which require very little or no additional data at all for calibration in two-dimensional multislice acquisitions. Instead of additional data, information from spatially adjacent slices is used to estimate coil sensitivity information, thereby increasing the efficiency of parallel imaging. The proposed approaches rely on the assumption that over a small range of slices, coil sensitivities vary smoothly in slice direction. Both methods are implemented as variants of the GRAPPA algorithm. For a given effective acceleration, superior image quality is achieved compared to the conventional GRAPPA method. For the latter calibration lines for coil weight computation must be acquired in addition to the undersampled k-spaces for coil weight computation, thus requiring higher k-space undersampling, that is, a higher reduction factor to achieve the same effective acceleration. The proposed methods are particularly suitable to speed up parallel imaging for clinical applications where the reduction factor is limited to two or three.

Regionally Optimized Reconstruction for Partially Parallel Imaging in MRI Applications

Based on the conventional SENSE and GRAPPA, a regionally optimized reconstruction method is developed for reduced noise and artifact level in partially parallel imaging. In this regionally optimized reconstruction, the field-of-view (FOV) is divided into a number of small regions. Over every small region, the noise amplification and data fitting error can be balanced and minimized locally by taking advantage of spatial correlation of neighboring pixels in reconstruction. The full FOV image can be obtained by "region-by-region" reconstruction. Compared with the conventional SENSE, this method gives better performance in the regions where there are pixels with high SENSE g-factors. Compared with GRAPPA, it is better in the regions where all the pixels have low SENSE g-factors. In this work, we applied the regionally optimized reconstruction in four important imaging experiments: brain, spine, breast, and cardiac. It was demonstrated in these experiments that the overall image quality using this regionally optimized reconstruction is better than that using the conventional SENSE or GRAPPA.

Spectral phase‐corrected GRAPPA reconstruction of three‐dimensional echo‐planar spectroscopic imaging (3D‐EPSI)

MR spectroscopic (MRS) images from a large volume of brain can be obtained using a 3D echo-planar spectroscopic imaging (3D-EPSI) sequence. However, routine applications of 3D-EPSI are still limited by a long scan time. In this communication, a new approach termed "spectral phase-corrected generalized autocalibrating partially parallel acquisitions" (SPC-GRAPPA) is introduced for the reconstruction of 3D-EPSI data to accelerate data acquisition while preserving the accuracy of quantitation of brain metabolites. In SPC-GRAPPA, voxel-by-voxel spectral phase alignment between metabolite 3D-EPSI from individual coil elements is performed in the frequency domain, utilizing the whole spectrum from interleaved water reference 3D-EPSI for robust estimation of the zero-order phase correction. The performance of SPC-GRAPPA was compared with that of fully encoded 3D-EPSI and conventional GRAPPA. Analysis of whole-brain 3D-EPSI data reconstructed by SPC-GRAPPA demonstrates that SPC-GRAPPA with an acceleration factor of 1.5 yields results very similar to those obtained by fully encoded 3D-EPSI, and is more accurate than conventional GRAPPA.