ENLIVE: An Efficient Nonlinear Method for Calibrationless and Robust Parallel Imaging

Purpose: We propose and evaluate a new structured low-rank method for echoplanar imaging (EPI) ghost correction called Robust Autocalibrated LORAKS (RAC-LORAKS). The method can be used to suppress EPI ghosts arising from the differences between different readout gradient polarities and/or the differences between different shots. It does not require conventional EPI navigator signals, and is robust to imperfect autocalibration data. Methods: Autocalibrated LORAKS is a previous structured low-rank method for EPI ghost correction that uses GRAPPA-type autocalibration data to enable highquality ghost correction. This method works well when the autocalibration data are pristine, but performance degrades substantially when the autocalibration information is imperfect. RAC-LORAKS generalizes Autocalibrated LORAKS in two ways. First, it does not completely trust the information from autocalibration data, and instead considers the autocalibration and EPI data simultaneously when estimating low-rank matrix structure. Second, it uses complementary information from the autocalibration data to improve EPI reconstruction in a multi-contrast joint reconstruction framework. RAC-LORAKS is evaluated using simulations and in vivo data, including comparisons to state-of-the-art methods. Results: RAC-LORAKS is demonstrated to have good ghost elimination performance compared to state-of-the-art methods in several complicated EPI acquisition scenarios (including gradient-echo brain imaging, diffusion-encoded brain imaging, and cardiac imaging). Conclusions: RAC-LORAKS provides effective suppression of EPI ghosts and is robust to imperfect autocalibration data.

Section: Introductionmentioning

confidence: 99%

Robust autocalibrated structured low‐rank EPI ghost correction

Lobos

Hoge

Javed

et al. 2020

“…Comparison of MPI (MPI‐BL and MPI‐L) with ESPIRiT 19 reconstruction using 2 maps and ENLIVE 7 reconstruction using 2 maps of a single slice of a human head, undersampled with Cartesian CAIPIRINHA patterns with differing undersampling factors R and fixed size of the ACS region (24). The MPI‐BL joint density (third column) and SMs reconstruction are obtained with a basis of dimension

q = 200

; the reconstructed SMs are subsequently used as an input for the MPI‐L density reconstruction (last column)…”

Section: Resultsmentioning

confidence: 99%

“…More specifically, in both methods, the authors exploit the smoothness of the SMs by making use of a polynomial expansion for constraining the subspace of the possible solutions of the SMs in the former while applying a smoothness‐enforcing regularization term in the later. Recently, NLINV was further generalized to a method dubbed ENLIVE with the addition of extra bilinear forms in order to account for the violation of the standard model in case of limited field‐of‐view (FOV) 7 . Another aspect of the smoothness and the spatial selectivity of the SMs is that they also favor purely algebraic techniques based on modern numerical linear algebra algorithms that promote low‐rank and subspace‐specific solutions 8‐11 …”

Section: Introductionmentioning

confidence: 99%

Maxwell parallel imaging

Francavilla¹,

Lefkimmiatis²,

Villena³

et al. 2021

Purpose To develop a general framework for parallel imaging (PI) with the use of Maxwell regularization for the estimation of the sensitivity maps (SMs) and constrained optimization for the parameter‐free image reconstruction. Theory and Methods Certain characteristics of both the SMs and the images are routinely used to regularize the otherwise ill‐posed optimization‐based joint reconstruction from highly accelerated PI data. In this paper, we rely on a fundamental property of SMs—they are solutions of Maxwell equations—we construct the subspace of all possible SM distributions supported in a given field‐of‐view, and we promote solutions of SMs that belong in this subspace. In addition, we propose a constrained optimization scheme for the image reconstruction, as a second step, once an accurate estimation of the SMs is available. The resulting method, dubbed Maxwell parallel imaging (MPI), works for both 2D and 3D, with Cartesian and radial trajectories, and minimal calibration signals. Results The effectiveness of MPI is illustrated for various undersampling schemes, including radial, variable‐density Poisson‐disc, and Cartesian, and is compared against the state‐of‐the‐art PI methods. Finally, we include some numerical experiments that demonstrate the memory footprint reduction of the constructed Maxwell basis with the help of tensor decomposition, thus allowing the use of MPI for full 3D image reconstructions. Conclusion The MPI framework provides a physics‐inspired optimization method for the accurate and efficient image reconstruction from arbitrary accelerated scans.

“…We performed an in vivo measurement on a human heart (short‐axis view, 30 channel thorax and spine coil, FLASH sequence, FOV = 256 × 256 mm, oversampled readout samples = 320, TE/TR = 1.47/2.3 ms, slice thickness = 8 mm) using a full circle golden angle acquisition scheme. Seventy‐five consecutive spokes during the end‐diastole were combined for image reconstruction with ENLIVE . The gradient delays were estimated using RING and the AC‐Adaptive method utilizing all numbers of spokes in the range

N_{sp} \in false[3, 75 false]

.…”

Section: Methodsmentioning

confidence: 99%

“…Seventy-five consecutive spokes during the end-diastole were combined for image reconstruction with ENLIVE. 36 The gradient delays were estimated using RING and the AC-Adaptive method utilizing all numbers of spokes in the range N sp ∈ [3,75]. The L2 errors were calculated as described previously.…”

Section: In Vivo Measurementsmentioning

confidence: 99%

Simple auto‐calibrated gradient delay estimation from few spokes using Radial Intersections (RING)

Rosenzweig

Holme

Uecker

2018

Self Cite

Purpose To develop a simple and robust tool for the estimation of gradient delays from highly undersampled radial k‐space data. Theory In radial imaging gradient delays induce parallel and orthogonal trajectory shifts, which can be described using an ellipse model. The intersection points of the radial spokes, which can be estimated by spoke‐by‐spoke comparison of k‐space samples, distinctly determine the parameters of the ellipse. Using the proposed method (RING), these parameters can be obtained using a least‐squares fit and utilized for the correction of gradient delays. Methods The functionality and accuracy of the proposed RING method is validated and compared to correlation‐based gradient‐delay estimation from opposing spokes using numerical simulations, phantom and in vivo heart measurements. Results In all experiments, RING robustly provides accurate gradient delay estimations even for as few as three radial spokes. Conclusions The simple and straightforward to implement RING method provides accurate gradient delay estimation for highly undersampled radial imaging.