Robust SENSE reconstruction of simultaneous multislice EPI with low‐rank enhanced coil sensitivity calibration and slice‐dependent 2D Nyquist ghost correction

Lyu, Mengye; Barth, Markus; Xie, Victor B.; Liu, Yilong; Ma, Xiaoyi; Feng, Yanqiu; Wu, Ed X.

doi:10.1002/mrm.27120

Cited by 17 publications

(45 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Another class of high‐performance reference‐free EPI ghost correction methods have been proposed using low‐rank matrix completion approaches . Specifically, Lee et al used the annihilating filter‐based low‐rank Hankel matrix approach (ALOHA), where the key idea is to take advantage of the fact that the concatenated Hankel matrix, which consists of even and odd k‐space lines, has a low‐rank structure due to the multi‐channel redundancy.…”

Section: Introductionmentioning

confidence: 99%

k‐Space deep learning for reference‐free EPI ghost correction

Lee

Han

Ryu

et al. 2019

Magnetic Resonance in Med

View full text Add to dashboard Cite

Purpose Nyquist ghost artifacts in echo planar imaging (EPI) are originated from phase mismatch between the even and odd echoes. However, conventional correction methods using reference scans often produce erroneous results especially in high‐field MRI due to the nonlinear and time‐varying local magnetic field changes. Recently, it was shown that the problem of ghost correction can be reformulated as k‐space interpolation problem that can be solved using structured low‐rank Hankel matrix approaches. Another recent work showed that data driven Hankel matrix decomposition can be reformulated to exhibit similar structures as deep convolutional neural network. By synergistically combining these findings, we propose a k‐space deep learning approach that immediately corrects the phase mismatch without a reference scan in both accelerated and non‐accelerated EPI acquisitions. Theory and Methods To take advantage of the even and odd‐phase directional redundancy, the k‐space data are divided into 2 channels configured with even and odd phase encodings. The redundancies between coils are also exploited by stacking the multi‐coil k‐space data into additional input channels. Then, our k‐space ghost correction network is trained to learn the interpolation kernel to estimate the missing virtual k‐space data. For the accelerated EPI data, the same neural network is trained to directly estimate the interpolation kernels for missing k‐space data from both ghost and subsampling. Results Reconstruction results using 3T and 7T in vivo data showed that the proposed method outperformed the image quality compared to the existing methods, and the computing time is much faster. Conclusions The proposed k‐space deep learning for EPI ghost correction is highly robust and fast, and can be combined with acceleration, so that it can be used as a promising correction tool for high‐field MRI without changing the current acquisition protocol.

show abstract

Section: Introductionmentioning

confidence: 99%

k‐Space deep learning for reference‐free EPI ghost correction

Lee

Han

Ryu

et al. 2019

Magnetic Resonance in Med

View full text Add to dashboard Cite

show abstract

“…Such inconsistency can be caused by various acquisition imperfections, such as gradient and ADC delay, eddy current, and field inhomogeneity. It manifests as Nyquist ghost artifacts in reconstructed EPI images . Regardless of the physical sources, Nyquist ghosts can be mathematically described by the phase difference between positive‐ and negative‐echo images, The estimation of this inherently 2D and nonlinear phase error map for Nyquist ghost correction has been attempted by using reference scans or navigator echoes at the expense of prolonged acquisition time.…”

Section: Introductionmentioning

confidence: 99%

“…This phase error‐correction SENSE (PEC‐SENSE) approach resembles that used in multiplexed sensitivity encoding for multi‐shot diffusion imaging . However, such parallel imaging‐based correction utilizes coil sensitivity maps, which themselves can be contaminated by 2D Nyquist ghosts if EPI‐based sequence is used for coil sensitivity calibration …”

Section: Introductionmentioning

confidence: 99%

“…Moreover, the 2D intershot phase variations within the single‐band multi‐shot EPI reference scan due to the physiological and hardware factors can introduce extra artifacts, further degrading the coil sensitivity calibration. Such degraded coil sensitivity calibration will undermine the subsequent SMS reconstruction with increased artifacts or decreased SNR …”

Section: Introductionmentioning

confidence: 99%

“…However, coil sensitivity map masking and empirical rank selection are necessary in the aforementioned PEC‐SENSE SMS EPI reconstruction, making it cumbersome for practical implementation. First, to minimize noise and artifact amplification, PEC‐SENSE approach needs a binary mask to exclude region outside the object from SENSE reconstruction by setting its coil sensitivity to zero. Conventionally, the binary mask can be obtained by thresholding over the reference images, and the threshold is set to order of the noise or portion of the maximum signal intensity .…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

PEC‐GRAPPA reconstruction of simultaneous multislice EPI with slice‐dependent 2D Nyquist ghost correction

Liu

Lyu

Barth

et al. 2018

Magnetic Resonance in Med

Self Cite

View full text Add to dashboard Cite

Purpose To provide simultaneous multislice (SMS) EPI reconstruction with k‐space implementation and robust Nyquist ghost correction. Methods 2D phase error correction SENSE (PEC‐SENSE) was recently developed for Nyquist ghost correction in SMS EPI reconstruction for which virtual coil simultaneous autocalibration and k‐space estimation (VC‐SAKE) was used to remove slice‐dependent Nyquist ghosts and intershot 2D phase variations in multi‐shot EPI reference scan. However, masking coil sensitivity maps to exclude background region in PEC‐SENSE and manually selecting slice‐wise target ranks in VC‐SAKE are cumbersome procedures in practice. To avoid masking, the concept of PEC‐SENSE is extended to k‐space implementation and termed as PEC‐GRAPPA. Furthermore, a singular value shrinkage scheme is incorporated in VC‐SAKE to circumvent the empirical slice‐wise target rank selection. PEC‐GRAPPA was evaluated and compared to PEC‐SENSE with/without masking and 1D linear phase correction GRAPPA. Results PEC‐GRAPPA robustly reconstructed SMS EPI images from 7T phantom and human brain data, effectively removing the phase error‐induced artifacts. The resulting residual artifact level and temporal SNR were comparable to those by PEC‐SENSE with careful tuning. PEC‐GRAPPA outperformed PEC‐SENSE without masking and 1D linear phase correction GRAPPA. Conclusion Our proposed PEC‐GRAPPA approach effectively removes the artifacts caused by Nyquist ghosts in SMS EPI without cumbersome tuning. This approach provides a robust and practical implementation of SMS EPI reconstruction in k‐space with slice‐dependent 2D Nyquist ghost correction.

show abstract

Learning from synthetic data for reference‐free Nyquist ghost correction and parallel imaging reconstruction of echo planar imaging

et al. 2022

View full text Add to dashboard Cite

Background: Echo planar imaging (EPI) suffers from Nyquist ghost caused by eddy currents and other non-ideal factors. Deep learning has received interest for EPI ghost correction.However,large datasets with qualified labels are usually unavailable, especially for the under-sampled EPI data due to the imperfection of traditional ghost correction algorithms. Purpose: To develop a multi-coil synthetic-data-based deep learning method for the Nyquist ghost correction and reconstruction of under-sampled EPI. Methods: Our network is trained purely with synthetic data. The labels of the training samples are generated by combining a public magnetic resonance imaging dataset and a few pre-collected coil sensitivity maps. The input is synthesized by under-sampling (for the accelerated case) and adding phase errors between the even and odd echoes of the label. To bridge the gap between synthetic data and data from real acquisition, linear and non-linear 2D phase errors are considered during the training data generation. Results:The proposed method outperformed the existing mainstream approaches in several experiments. The average ghost-to-signal ratios of our/3-line navigator-based methods were 0.51%/5.36% and 0.42%/8.64% in fully-sampled and under-sampled in vivo experiments, respectively. In the sagittal experiments, our method successfully corrected higher-order and 2D phase errors. Our method also outperformed other reference-based methods on motion-corrupted data. In the simulation experiments, the peak signal-to-noise ratios were 37.6/38.3 dB for 2D linear/non-linear simulated phase errors, indicating that our method was consistently reliable for different kinds of phase errors. Conclusion: Our method achieves superb ghost correction and parallel imaging reconstruction without any calibration information, and can be readily adapted to other EPI-based applications.

show abstract

Robust SENSE reconstruction of simultaneous multislice EPI with low‐rank enhanced coil sensitivity calibration and slice‐dependent 2D Nyquist ghost correction

Cited by 17 publications

References 41 publications

k‐Space deep learning for reference‐free EPI ghost correction

k‐Space deep learning for reference‐free EPI ghost correction

PEC‐GRAPPA reconstruction of simultaneous multislice EPI with slice‐dependent 2D Nyquist ghost correction

Learning from synthetic data for reference‐free Nyquist ghost correction and parallel imaging reconstruction of echo planar imaging

Contact Info

Product

Resources

About