Recurrent Variational Network: A Deep Learning Inverse Problem Solver applied to the task of Accelerated MRI Reconstruction

Yiasemis, George; Sánchez, Clara I.; Sonke, J.J.; Teuwen, Jonas

doi:10.1109/cvpr52688.2022.00081

Cited by 36 publications

(16 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A common method for estimating them is by fully-sampling a small region of the center of the k-space, also known as the autocalibration signal (ACS) which includes low frequencies [11,10].…”

Section: Sensitivity Map Estimationmentioning

confidence: 99%

“…Optimization of Eq. 17 may alternatively be performed in the k-space domain as demonstrated by some authors [10,11]:…”

Section: Deep Learning-based Accelerated Mri Reconstructionmentioning

confidence: 99%

“…With the recent advancements in Deep Learning (DL) and Computer Vision (CV), a plethora of algorithms have emerged targeting to solve imaging inverse problems, with Accelerated MRI Reconstruction being a par excellence example. Combined with CS and PI, numerous DL-based methods involving convolutional neural networks (CNNs) have been proposed in the literature [10,11,12,13,14,15,16,17] applied to the task of Accelerated MRI Reconstruction. These methods are usually trained in a supervised manner using retrospectively subsampled (from available fully-sampled) k-space datasets and their target is to make a prediction of the fully-sampled k-space or its image reconstruction.…”

Section: Introductionmentioning

confidence: 99%

“…In this work, we aim to investigate and compare the effects of employing miscellaneous retrospective subsampling schemes on the quality of DL-based learned image reconstructions. To that end, we trained and tested Recurrent Variational Networks [10] (RecurrentVarNets) on retrospectively subsampled k-space measurements. We performed experiments under either scheme-specific or multi-scheme setups, in which models were trained and evaluated on data subsampled with either individual and multiple, respectively, subsampling schemes.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

On Retrospective k-space Subsampling schemes For Deep MRI Reconstruction

Yiasemis¹,

Sánchez²,

Sonke³

et al. 2023

Preprint

View full text Add to dashboard Cite

Purpose: The MRI k-space acquisition is time consuming. Traditional techniques aim to acquire accelerated data, which in conjunction with recent DL methods, aid in producing high-fidelity images in truncated times. Conventionally, subsampling the k-space is performed by utilizing Cartesianrectilinear trajectories, which even with the use of DL, provide imprecise reconstructions, though, a plethora of non-rectilinear or non-Cartesian trajectories can be implemented in modern MRI scanners. This work investigates the effect of the k-space subsampling scheme on the quality of reconstructed accelerated MRI measurements produced by trained DL models. Methods:The RecurrentVarNet was used as the DL-based MRI-reconstruction architecture. Cartesian fully-sampled multi-coil k-space measurements from three datasets with different accelerations were retrospectively subsampled using eight distinct subsampling schemes (four Cartesian-rectilinear, two Cartesian non-rectilinear, two non-Cartesian). Experiments were conducted in two frameworks: Scheme-specific, where a distinct model was trained and evaluated for each dataset-subsampling scheme pair, and multi-scheme, where for each dataset a single model was trained on data randomly subsampled by any of the eight schemes and evaluated on data subsampled by all schemes.Results: In the scheme-specific setting RecurrentVarNets trained and evaluated on non-rectilinearly subsampled data demonstrated superior performance especially for high accelerations, whilst in the multi-scheme setting, reconstruction performance on rectilinearly subsampled data improved when compared to the scheme-specific experiments. Conclusion:Training DL-based MRI reconstruction algorithms on non-rectilinearly subsampled measurements can produce more faithful reconstructions. Our findings demonstrate the potential for using DL-based methods trained on prospective acquisitions with non-rectilinearly subsampled measurements to optimize scan time and image quality.

show abstract

Section: Sensitivity Map Estimationmentioning

confidence: 99%

“…Optimization of Eq. 17 may alternatively be performed in the k-space domain as demonstrated by some authors [10,11]:…”

Section: Deep Learning-based Accelerated Mri Reconstructionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On Retrospective k-space Subsampling schemes For Deep MRI Reconstruction

Yiasemis¹,

Sánchez²,

Sonke³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…Traditionally, optimization-based methods for joint image reconstruction and CSM estimation treat S as another unknown variable in (3) and alternate between updating the image and updating the coil sensitivities [43,44]. Deep unfolding has recently been adopted to perform joint estimation of image and CSMs without any pre-calibration procedure [18][19][20]46]. The concept behind these methods is to model CSM estimated as a trainable DNN module that can be optimized simultaneously with other learnable parameters in the deep network.…”

Section: Joint Reconstruction and Csm Estimationmentioning

confidence: 99%

SPICE: Self-Supervised Learning for MRI with Automatic Coil Sensitivity Estimation

Hu¹,

Gan²,

Ying³

et al. 2022

Preprint

View full text Add to dashboard Cite

Deep model-based architectures (DMBAs) integrating physical measurement models and learned image regularizers are widely used in parallel magnetic resonance imaging (PMRI). Traditional DMBAs for PMRI rely on preestimated coil sensitivity maps (CSMs) as a component of the measurement model. However, estimation of accurate CSMs is a challenging problem when measurements are highly undersampled. Additionally, traditional training of DMBAs requires high-quality groundtruth images, limiting their use in applications where groundtruth is difficult to obtain. This paper addresses these issues by presenting SPICE as a new method that integrates self-supervised learning and automatic coil sensitivity estimation. Instead of using pre-estimated CSMs, SPICE simultaneously reconstructs accurate MR images and estimates high-quality CSMs. SPICE also enables learning from undersampled noisy measurements without any groundtruth. We validate SPICE on experimentally collected data, showing that it can achieve state-of-the-art performance in highly accelerated data acquisition settings (up to 10×).

show abstract

Reconstruction of Cardiac Cine MRI Using Motion‐Guided Deformable Alignment and Multi‐Resolution Fusion

Han,

Chen,

Liu

et al. 2024

Int J Imaging Syst Tech

View full text Add to dashboard Cite

Cardiac cine magnetic resonance imaging (MRI) is one of the important means to assess cardiac functions and vascular abnormalities. Mitigating artifacts arising during image reconstruction and accelerating cardiac cine MRI acquisition to obtain high‐quality images is important. A novel end‐to‐end deep learning network is developed to improve cardiac cine MRI reconstruction. First, a U‐Net is adopted to obtain the initial reconstructed images in k‐space. Further to remove the motion artifacts, the motion‐guided deformable alignment (MGDA) module with second‐order bidirectional propagation is introduced to align the adjacent cine MRI frames by maximizing spatial–temporal information to alleviate motion artifacts. Finally, the multi‐resolution fusion (MRF) module is designed to correct the blur and artifacts generated from alignment operation and obtain the last high‐quality reconstructed cardiac images. At an 8× acceleration rate, the numerical measurements on the ACDC dataset are structural similarity index (SSIM) of 78.40% ± 4.57%, peak signal‐to‐noise ratio (PSNR) of 30.46 ± 1.22 dB, and normalized mean squared error (NMSE) of 0.0468 ± 0.0075. On the ACMRI dataset, the results are SSIM of 87.65% ± 4.20%, PSNR of 30.04 ± 1.18 dB, and NMSE of 0.0473 ± 0.0072. The proposed method exhibits high‐quality results with richer details and fewer artifacts for cardiac cine MRI reconstruction on different accelerations.

show abstract

Recurrent Variational Network: A Deep Learning Inverse Problem Solver applied to the task of Accelerated MRI Reconstruction

Cited by 36 publications

References 28 publications

On Retrospective k-space Subsampling schemes For Deep MRI Reconstruction

On Retrospective k-space Subsampling schemes For Deep MRI Reconstruction

SPICE: Self-Supervised Learning for MRI with Automatic Coil Sensitivity Estimation

Reconstruction of Cardiac Cine MRI Using Motion‐Guided Deformable Alignment and Multi‐Resolution Fusion

Contact Info

Product

Resources

About