A Deep Information Sharing Network for Multi-Contrast Compressed Sensing MRI Reconstruction

Sun, Liankun; Fan, Zhiwen; Fu, Xueyang; Huang, Yue; Ding, Xinghao; Paisley, John

doi:10.1109/tip.2019.2925288

Cited by 95 publications

(63 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…We believe that this can be seamlessly incorporated with the other acceleration methods, such as CS . Compressed sensing and ANN have been used for acceleration of MRI in general, but ANN‐based approaches often outperform compressed sensing methods …”

Section: Discussionmentioning

confidence: 99%

“…9 Compressed sensing and ANN have been used for acceleration of MRI in general, but ANN-based approaches often outperform compressed sensing methods. 11,13,17,41,42 For the validity of retrospective acquisition of low SEMAC factor data from high SEMAC factor data, SEMAC F I G U R E 5 Visual comparison of CNN, Parallel Imaging, and Compressed sensing reconstruction results. Each row shows slices of a representative phantom.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Artificial neural network for Slice Encoding for Metal Artifact Correction (SEMAC) MRI

Seo

Joon

Luu

et al. 2019

Magnetic Resonance in Med

View full text Add to dashboard Cite

Purpose To develop new artificial neural networks (ANNs) to accelerate slice encoding for metal artifact correction (SEMAC) MRI. Methods Eight titanium phantoms and 77 patients after brain tumor surgery involving metallic neuro‐plating instruments were scanned using SEMAC at a 3T Skyra scanner. For the phantoms, proton‐density, T1‐, and T2‐weighted images were acquired for developing both multilayer perceptron (MLP) and convolutional neural network (CNN). For the patients, T2‐weighted images were acquired for developing CNN. All networks were trained with the SEMAC factor 4 or 6 as input and the factor 12 as label, yielding an acceleration factor of 3 or 2. Performance of the CNN model was compared against parallel imaging and compressed sensing on the phantom datasets. Two extra T1‐weighted in vivo sets were acquired to investigate generalizability of the models to different contrasts. Results Both multilayer perceptron and CNN provided artifact‐suppressed images better than the input images and comparable to the label images visually and quantitatively, a trend observable regardless of input SEMAC factor and image type (P < .01). CNN suppressed the artifacts better than multilayer perceptron, parallel imaging, and compressed sensing (P < .01). Tests on the patient datasets demonstrated clear metal artifact suppression visually and quantitatively (P < .01). Tests on T1 datasets also demonstrated clear visual metal artifact suppression. Conclusion Our study introduced a new effective way of artificial neural networks to accelerate SEMAC MRI while maintaining the comparable quality of metal artifact suppression. Application on the preliminary patient datasets proved the feasibility in clinical usage, which warrants further investigation.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Artificial neural network for Slice Encoding for Metal Artifact Correction (SEMAC) MRI

Seo

Joon

Luu

et al. 2019

Magnetic Resonance in Med

View full text Add to dashboard Cite

show abstract

“…Deep neural networks were previously proposed for recovery of multi-contrast MR acquisitions where each acquisition was accelerated at an identical rate [53]- [55]. Despite improved recovery compared to isolated reconstruction of individual contrasts, joint reconstruction may still suffer from loss of high-spatial-frequency information towards higher acceleration factors.…”

Section: Introductionmentioning

confidence: 99%

Prior-Guided Image Reconstruction for Accelerated Multi-Contrast MRI via Generative Adversarial Networks

Dar

Yurt

Shahdloo

et al. 2020

IEEE J. Sel. Top. Signal Process.

View full text Add to dashboard Cite

Multi-contrast MRI acquisitions of an anatomy enrich the magnitude of information available for diagnosis. Yet, excessive scan times associated with additional contrasts may be a limiting factor. Two mainstream frameworks for enhanced scan efficiency are reconstruction of undersampled acquisitions and synthesis of missing acquisitions. Recently, deep learning methods have enabled significant performance improvements in both frameworks. Yet, reconstruction performance decreases towards higher acceleration factors with diminished sampling density at high-spatial-frequencies, whereas synthesis can manifest artefactual sensitivity or insensitivity to image features due to the absence of data samples from the target contrast. Here we propose a new approach for synergistic recovery of undersampled multi-contrast acquisitions based on conditional generative adversarial networks. The proposed method mitigates the limitations of pure learning-based reconstruction or synthesis by utilizing three priors: shared high-frequency prior available in the source contrast to preserve high-spatial-frequency details, low-frequency prior available in the undersampled target contrast to prevent feature leakage/loss, and perceptual prior to improve recovery of high-level features. Demonstrations on brain MRI datasets from healthy subjects and patients indicate the superior performance of the proposed method compared to pure reconstruction and synthesis methods. The proposed method can help improve the quality and scan efficiency of multi-contrast MRI exams.

show abstract

“…There have been attempts to reconstruct fully sampled multicontrast images from down-sampled multicontrast images using deeplearning or compressed-sensing algorithms. [14][15][16][17][18] However, these methods can be detrimentally affected by motion which occurs between scans such that they require additional registration processes between images.…”

Section: Introductionmentioning

confidence: 99%

Reconstruction of multicontrast MR images through deep learning

et al. 2020

View full text Add to dashboard Cite

Purpose Magnetic resonance (MR) imaging with a long scan time can lead to degraded images due to patient motion, patient discomfort, and increased costs. For these reasons, the role of rapid MR imaging is important. In this study, we propose the joint reconstruction of multicontrast brain MR images from down‐sampled data to accelerate the data acquisition process using a novel deep‐learning network. Methods Twenty‐one healthy volunteers (female/male = 7/14, age = 26 ± 4 yr, range 22–35 yr) and 16 postoperative patients (female/male = 7/9, age = 49 ± 9 yr, range 37–62 yr) were scanned on a 3T whole‐body scanner for prospective and retrospective studies, respectively, using both T1‐weighted spin‐echo (SE) and T2‐weighted fast spin‐echo (FSE) sequences. We proposed a network which we term “X‐net” to reconstruct both T1‐ and T2‐weighted images from down‐sampled images as well as a network termed “Y‐net” which reconstructs T2‐weighted images from highly down‐sampled T2‐weighted images and fully sampled T1‐weighted images. Both X‐net and Y‐net are composed of two concatenated subnetworks. We investigate optimal sampling patterns, the optimal patch size for augmentation, and the optimal acceleration factors for network training. An additional Y‐net combined with a generative adversarial network (GAN) was also implemented and tested to investigate the effects of the GAN on the Y‐net performance. Single‐ and joint‐reconstruction parallel‐imaging and compressed‐sensing algorithms along with a conventional U‐net were also tested and compared with the proposed networks. For this comparison, the structural similarity (SSIM), normalized mean square error (NMSE), and Fréchet inception distance (FID) were calculated between the outputs of the networks and fully sampled images. The statistical significance of the performance was evaluated by assessing the interclass correlation and in paired t‐tests. Results The outputs from the two concatenated subnetworks were closer to the fully sampled images compared to those from one subnetwork, with this result showing statistical significance. Uniform down‐sampling led to a statically significant improvement in the image quality compared to random or central down‐sampling patterns. In addition, the proposed networks provided higher SSIM and NMSE values than U‐net, compressed‐sensing, and parallel‐imaging algorithms, all at statistically significant levels. The GAN‐based Y‐net showed a better FID and more realistic images compared to a non‐GAN‐based Y‐net. The performance capabilities of the networks were similar between normal subjects and patients. Conclusions The proposed X‐net and Y‐net effectively reconstructed full images from down‐sampled images, outperforming the conventional parallel‐imaging, compressed‐sensing and U‐net methods and providing more realistic images in combination with a GAN. The developed networks potentially enable us to accelerate multicontrast anatomical MR imaging in routine clinical studies including T1‐and T2‐weighted imaging.

show abstract

A Deep Information Sharing Network for Multi-Contrast Compressed Sensing MRI Reconstruction

Cited by 95 publications

References 33 publications

Artificial neural network for Slice Encoding for Metal Artifact Correction (SEMAC) MRI

Artificial neural network for Slice Encoding for Metal Artifact Correction (SEMAC) MRI

Prior-Guided Image Reconstruction for Accelerated Multi-Contrast MRI via Generative Adversarial Networks

Reconstruction of multicontrast MR images through deep learning

Contact Info

Product

Resources

About