Deep Magnetic Resonance Image Reconstruction: Inverse Problems Meet Neural Networks

Liang, Dong; Cheng, Jing; Ke, Ziwen; Ying, Leslie

doi:10.1109/msp.2019.2950557

Cited by 287 publications

(196 citation statements)

References 38 publications

Supporting

Mentioning

180

Contrasting

Unclassified

Order By: Relevance

“…In the same image‐domain reconstruction setting, a self‐supervised learning scheme using cycleGANs with optimal transport cost minimization was proposed, 66 although initial results exhibit blurring artifacts. Although purely data‐driven image domain methods have been used for DL‐MRI reconstruction, physics‐guided DL‐MRI techniques are more desirable, as they offer a degree of interpretability by incorporating domain knowledge on the MRI encoding mechanism 20,27,28,30,31,33 . In this physics‐guided setting, an earlier work used the output of a regularized CG‐SENSE algorithm based on compressed sensing as the reference for supervised training, 67 showing that such training may outperform the compressed‐sensing output, as some images are overregularized while others are underregularized.…”

Section: Discussionmentioning

confidence: 99%

“…Recently, deep learning (DL) has gained interest for high‐quality accelerated MRI. Deep learning–based MRI reconstruction algorithms can be roughly divided into two categories: purely data‐driven and physics‐guided 20 . In purely data‐driven approaches, a mapping between the undersampled k‐space/aliased image to full k‐space/artifact‐free image is learned 21‐26 .…”

Section: Introductionmentioning

confidence: 99%

“…The unrolled network alternates between data consistency (DC)and regularization, in which the regularization is implemented implicitly using a neural network. Subsequently, these unrolled networks are trained end‐to‐end with a loss function that characterizes similarity with a reference image obtained from fully sampled data 20 . The parameters of the network can be different across the unrolled iterations 27,31 or shared across them 28,33 …”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Self‐supervised learning of physics‐guided reconstruction neural networks without fully sampled reference data

Yaman

Hosseini

Moeller

et al. 2020

Magnetic Resonance in Med

181

227

View full text Add to dashboard Cite

Purpose To develop a strategy for training a physics‐guided MRI reconstruction neural network without a database of fully sampled data sets. Methods Self‐supervised learning via data undersampling (SSDU) for physics‐guided deep learning reconstruction partitions available measurements into two disjoint sets, one of which is used in the data consistency (DC) units in the unrolled network and the other is used to define the loss for training. The proposed training without fully sampled data is compared with fully supervised training with ground‐truth data, as well as conventional compressed‐sensing and parallel imaging methods using the publicly available fastMRI knee database. The same physics‐guided neural network is used for both proposed SSDU and supervised training. The SSDU training is also applied to prospectively two‐fold accelerated high‐resolution brain data sets at different acceleration rates, and compared with parallel imaging. Results Results on five different knee sequences at an acceleration rate of 4 shows that the proposed self‐supervised approach performs closely with supervised learning, while significantly outperforming conventional compressed‐sensing and parallel imaging, as characterized by quantitative metrics and a clinical reader study. The results on prospectively subsampled brain data sets, in which supervised learning cannot be used due to lack of ground‐truth reference, show that the proposed self‐supervised approach successfully performs reconstruction at high acceleration rates (4, 6, and 8). Image readings indicate improved visual reconstruction quality with the proposed approach compared with parallel imaging at acquisition acceleration. Conclusion The proposed SSDU approach allows training of physics‐guided deep learning MRI reconstruction without fully sampled data, while achieving comparable results with supervised deep learning MRI trained on fully sampled data.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Self‐supervised learning of physics‐guided reconstruction neural networks without fully sampled reference data

Yaman

Hosseini

Moeller

et al. 2020

Magnetic Resonance in Med

181

227

View full text Add to dashboard Cite

show abstract

“…They employed an offline CNN to realize the mapping of undersampled zero-filled MRI and fully sampled kspace data and achieved good reconstruction effectively. Deep learning based MRI reconstruction methods can be roughly divided into unrolling-based approaches and those not based on unrolling (Liang et al, 2020). Among them, the unrolling-based method usually expands the CS-based iterative reconstruction algorithm into a deep network, so that the parameters in the reconstruction algorithm can be learned by the network.…”

Section: Introductionmentioning

confidence: 99%

SARA-GAN: Self-Attention and Relative Average Discriminator Based Generative Adversarial Networks for Fast Compressed Sensing MRI Reconstruction

Yuan

Jiang

Wang

et al. 2020

Front. Neuroinform.

View full text Add to dashboard Cite

Research on undersampled magnetic resonance image (MRI) reconstruction can increase the speed of MRI imaging and reduce patient suffering. In this paper, an undersampled MRI reconstruction method based on Generative Adversarial Networks with the Self-Attention mechanism and the Relative Average discriminator (SARA-GAN) is proposed. In our SARA-GAN, the relative average discriminator theory is applied to make full use of the prior knowledge, in which half of the input data of the discriminator is true and half is fake. At the same time, a self-attention mechanism is incorporated into the high-layer of the generator to build long-range dependence of the image, which can overcome the problem of limited convolution kernel size. Besides, spectral normalization is employed to stabilize the training process. Compared with three widely used GAN-based MRI reconstruction methods, i.e., DAGAN, DAWGAN, and DAWGAN-GP, the proposed method can obtain a higher peak signal-to-noise ratio (PSNR) and structural similarity index measure(SSIM), and the details of the reconstructed image are more abundant and more realistic for further clinical scrutinization and diagnostic tasks.

show abstract

“…In this study, we compare the VN, trained with real and synthetically generated knee cartilage images, against CS approaches for mono and biexponential T 1ρ mapping 11,12 . It is not our intention here to compare different deep learning methods for image reconstruction 13,15,19,20 , but compare one good representative of this class against one good representative of CS, which is among the current state-of-the-art methods for T 1ρ mapping 11,12,21 . In order to have a fair comparison between these approaches, the VN and CS used the same pre-available data for training or tuning the parameters of the algorithms.…”

mentioning

confidence: 99%

Rapid mono and biexponential 3D-T1ρ mapping of knee cartilage using variational networks

Zibetti

Johnson

Sharafi

et al. 2020

Sci Rep

View full text Add to dashboard Cite

In this study we use undersampled MRI acquisition methods to obtain accelerated 3D mono and biexponential spin–lattice relaxation time in the rotating frame (T1ρ) mapping of knee cartilage, reducing the usual long scan time. We compare the accelerated T1ρ maps obtained by deep learning-based variational network (VN) and compressed sensing (CS). Both methods were compared with spatial (S) and spatio-temporal (ST) filters. Complex-valued fitting was used for T1ρ parameters estimation. We tested with seven in vivo and six synthetic datasets, with acceleration factors (AF) from 2 to 10. Median normalized absolute deviation (MNAD), analysis of variance (ANOVA), and coefficient of variation (CV) were used for analysis. The methods CS-ST, VN-S, and VN-ST performed well for accelerating monoexponential T1ρ mapping, with MNAD around 5% for AF = 2, which increases almost linearly with the AF to an MNAD of 13% for AF = 8, with all methods. For biexponential mapping, the VN-ST was the best method starting with MNAD of 7.4% for AF = 2 and reaching MNAD of 13.1% for AF = 8. The VN was able to produce 3D-T1ρ mapping of knee cartilage with lower error than CS. The best results were obtained by VN-ST, improving CS-ST method by nearly 7.5%.

show abstract

Deep Magnetic Resonance Image Reconstruction: Inverse Problems Meet Neural Networks

Cited by 287 publications

References 38 publications

Self‐supervised learning of physics‐guided reconstruction neural networks without fully sampled reference data

Self‐supervised learning of physics‐guided reconstruction neural networks without fully sampled reference data

SARA-GAN: Self-Attention and Relative Average Discriminator Based Generative Adversarial Networks for Fast Compressed Sensing MRI Reconstruction

Rapid mono and biexponential 3D-T1ρ mapping of knee cartilage using variational networks

Contact Info

Product

Resources

About