546 Hits in 5.0 sec

Structure Preserving Compressive Sensing MRI Reconstruction using Generative Adversarial Networks [article]

Puneesh Deora, Bhavya Vasudeva, Saumik Bhattacharya, Pyari Mohan Pradhan
2020 arXiv   pre-print
In this work, a novel generative adversarial network (GAN) based framework for CS-MRI reconstruction is proposed.  ...  Compressive sensing magnetic resonance imaging (CS-MRI) accelerates the acquisition of MR images by breaking the Nyquist sampling limit.  ...  Compressive sensing (CS) [10] can be used to accelerate the MRI acquisition process by undersampling the kspace data. Reconstruction of CS-MRI is an ill-posed inverse problem [13] .  ... 
arXiv:1910.06067v2 fatcat:xbbwkydjnjgohihnlxiteqdoca

Measuring Robustness in Deep Learning Based Compressive Sensing [article]

Mohammad Zalbagi Darestani, Akshay S. Chaudhari, Reinhard Heckel
2021 arXiv   pre-print
Deep neural networks give state-of-the-art accuracy for reconstructing images from few and noisy measurements, a problem arising for example in accelerated magnetic resonance imaging (MRI).  ...  However, recent works have raised concerns that deep-learning-based image reconstruction methods are sensitive to perturbations and are less robust than traditional methods: Neural networks (i) may be  ...  “Sparse MRI: The application of compressed sensing for rapid MR imaging”.  ... 
arXiv:2102.06103v2 fatcat:57p7hrq2abeqbb4x2ufntj2274

On Instabilities of Conventional Multi-Coil MRI Reconstruction to Small Adverserial Perturbations [article]

Chi Zhang, Jinghan Jia, Burhaneddin Yaman, Steen Moeller, Sijia Liu, Mingyi Hong, Mehmet Akçakaya
2021 arXiv   pre-print
Although deep learning (DL) has received much attention in accelerated MRI, recent studies suggest small perturbations may lead to instabilities in DL-based reconstructions, leading to concern for their  ...  We investigate instabilities caused by small adversarial attacks for multi-coil acquisitions.  ...  Several follow-up studies 10, 11 explored adversarial training frameworks to improve the robustness of DL-MRI reconstruction.  ... 
arXiv:2102.13066v1 fatcat:mig3afgwpjdn7dswevijcn7ze4

A Review of Image Compressed Sensing in Deep Learning

Kaiguo Xia, Lei Hu, Pengqiang Mao
2020 DEStech Transactions on Engineering and Technology Research  
This paper sorts out the current image compressed sensing reconstruction methods based on deep learning, analyzes the characteristics and key steps of the algorithm according to three different deep network  ...  reduces the time consuming, showing the great potential of deep learning in the field of compressed sensing reconstruction.  ...  Compressed sensing reconstruction based on Generative Adversarial Network Generative Adversarial Networks [12] was firstly proposed by Ian Goodfellow in 2014 , and now it is the best generative model  ... 
doi:10.12783/dtetr/mcaee2020/35015 fatcat:76cw4vfu7fcafmxtfeqv73kwmq

Motion Corrected Multishot MRI Reconstruction Using Generative Networks with Sensitivity Encoding [article]

Muhammad Usman, Muhammad Umar Farooq, Siddique Latif, Muhammad Asim,, Junaid Qadir
2019 arXiv   pre-print
The proposed framework first employs CG-SENSE reconstruction to produce the motion-corrupted image and then a generative adversarial network (GAN) is used to correct the motion artifacts.  ...  In this paper, we propose a novel generative networks based conjugate gradient SENSE (CG-SENSE) reconstruction framework for motion correction in multishot MRI.  ...  Motion-corrupted k-space data has been reconstructed using CG-SENSE (without motion correction) and then fed to the adversarial network, which is tasked to generate the motion-free images.  ... 
arXiv:1902.07430v5 fatcat:hn5cbad36rfuziiu2qjusvq77a

Compressed Sensing: From Research to Clinical Practice with Data-Driven Learning [article]

Joseph Y. Cheng, Feiyu Chen, Christopher Sandino, Morteza Mardani, John M. Pauly, Shreyas S. Vasanawala
2019 arXiv   pre-print
Compressed sensing in MRI enables high subsampling factors while maintaining diagnostic image quality. This technique enables shortened scan durations and/or improved image resolution.  ...  As a result, compressed sensing can have greater clinical impact.  ...  Generative adversarial networks (GANs) [26] can be used to model the properties of the ground truth images and to exploit that information for improving the reconstruction quality.  ... 
arXiv:1903.07824v1 fatcat:hizgf5xjkjab5ni5thzihr466m

2020 Index IEEE Transactions on Computational Imaging Vol. 6

2020 IEEE Transactions on Computational Imaging  
Chen, H., +, TCI 2020 276-290 IFR-Net: Iterative Feature Refinement Network for Compressed Sensing MRI.  ...  Allain, P., +, TCI 2020 109- 124 IFR-Net: Iterative Feature Refinement Network for Compressed Sensing MRI.  ... 
doi:10.1109/tci.2021.3054596 fatcat:puij7ztll5ai7alxrmqzsupcny

SANTIS: Sampling-Augmented Neural neTwork with Incoherent Structure for MR image reconstruction [article]

Fang Liu, Lihua Chen, Richard Kijowski, Li Feng
2018 arXiv   pre-print
SANTIS uses a data cycle-consistent adversarial network combining efficient end-to-end convolutional neural network mapping, data fidelity enforcement and adversarial training for reconstructing accelerated  ...  The undersampled images are generated by a fixed undersampling pattern in the training, and the trained network is then applied to reconstruct new images acquired with the same pattern in the inference  ...  The encoder is used to achieve efficient data compression while probing robust and spatial invariant image features of input images.  ... 
arXiv:1812.03278v1 fatcat:ltdhwqspyzc3pctwlhccbcuibu

Deep Attentive Wasserstein Generative Adversarial Networks for MRI Reconstruction with Recurrent Context-Awareness [article]

Yifeng Guo, Chengjia Wang, Heye Zhang, Guang Yang
2020 arXiv   pre-print
The performance of traditional compressive sensing-based MRI (CS-MRI) reconstruction is affected by its slow iterative procedure and noise-induced artefacts.  ...  In this study, we propose a new deep learning-based CS-MRI reconstruction method to fully utilise the relationship among sequential MRI slices by coupling Wasserstein Generative Adversarial Networks (WGAN  ...  Introduction Compressed sensing magnetic resonance imaging (CS-MRI) [1] has been proposed for accelerating MRI process.  ... 
arXiv:2006.12915v1 fatcat:uffx27v7zrae7l2riwybxyv34y

Neural Network-based Reconstruction in Compressed Sensing MRI Without Fully-sampled Training Data [article]

Alan Q. Wang, Adrian V. Dalca, Mert R. Sabuncu
2020 arXiv   pre-print
Compressed Sensing MRI (CS-MRI) has shown promise in reconstructing under-sampled MR images, offering the potential to reduce scan times.  ...  In this paper, we explore a novel strategy to train an unrolled reconstruction network in an unsupervised fashion by adopting a loss function widely-used in classical optimization schemes.  ...  Introduction Magnetic resonance (MR) imaging can be accelerated via under-sampling kspace -a technique known as Compressed Sensing MRI (CS-MRI) [25] .  ... 
arXiv:2007.14979v1 fatcat:qfajr23iyfgszkkjyhc24lynle

Generative Model Adversarial Training for Deep Compressed Sensing [article]

Ashkan Esmaeili
2021 arXiv   pre-print
In this work, we propound how to design such a low-to-high dimensional deep learning-based generator suiting for compressed sensing, while satisfying robustness to universal adversarial perturbations in  ...  Experiments on real-world datasets are provided to substantiate the efficacy of the proposed generative model adversarial training for deep compressed sensing.  ...  There exists certain works which address the robust deep compressed sensing for defending against adversarial attack.  ... 
arXiv:2106.10696v1 fatcat:6gbazsqgobfdhg3f4rndokh4ky

Adversarial and Perceptual Refinement for Compressed Sensing MRI Reconstruction [article]

Maximilian Seitzer and Guang Yang and Jo Schlemper and Ozan Oktay and Tobias Würfl and Vincent Christlein and Tom Wong and Raad Mohiaddin and David Firmin and Jennifer Keegan and Daniel Rueckert and Andreas Maier
2018 arXiv   pre-print
Deep learning approaches have shown promising performance for compressed sensing-based Magnetic Resonance Imaging.  ...  In this work, we propose a hybrid method, in which a visual refinement component is learnt on top of an MSE loss-based reconstruction network.  ...  Introduction Compressed sensing-based Magnetic Resonance Imaging (CS-MRI) is a promising paradigm allowing to accelerate MRI acquisition by reconstructing images from only a fraction of the normally required  ... 
arXiv:1806.11216v1 fatcat:fxthvsbphramrk5ajv3zo7c5xu

Deep De-Aliasing for Fast Compressive Sensing MRI [article]

Simiao Yu, Hao Dong, Guang Yang, Greg Slabaugh, Pier Luigi Dragotti, Xujiong Ye, Fangde Liu, Simon Arridge, Jennifer Keegan, David Firmin, Yike Guo
2017 arXiv   pre-print
Compressive Sensing (CS) theory has been perfectly matched to the MRI scanning sequence design with much less required raw data for the image reconstruction.  ...  Inspired by recent advances in deep learning for solving various inverse problems, we propose a conditional Generative Adversarial Networks-based deep learning framework for de-aliasing and reconstructing  ...  In this study, we proposed a novel conditional Generative Adversarial Networks (GAN) based deep learning architecture for fast CS-MRI.  ... 
arXiv:1705.07137v1 fatcat:jbd25mxmjfggvb6sdaqbavut2m

Retrospective Motion Correction in Multishot MRI using Generative Adversarial Network

Muhammad Usman, Siddique Latif, Muhammad Asim, Byoung-Dai Lee, Junaid Qadir
2020 Scientific Reports  
In this paper, we propose a novel generative adversarial network (GAN)-based conjugate gradient SENSE (CG-SENSE) reconstruction framework for motion correction in multishot MRI.  ...  First CG-SENSE reconstruction is employed to reconstruct an image from the motion-corrupted k-space data and then the GAN-based proposed framework is applied to correct the motion artifacts.  ...  MRI, where CG-SENSE is used to reconstruct motion-corrupted images, and the generator network of the GAN, in conjunction with the discriminator network, is tasked with motion correction (Figure Credit  ... 
doi:10.1038/s41598-020-61705-9 pmid:32179823 fatcat:t6ed7d4ebjf2ncg6akw7u5vmiy

DAGAN: Deep De-Aliasing Generative Adversarial Networks for Fast Compressed Sensing MRI Reconstruction

Guang Yang, Simiao Yu, Hao Dong, Greg Slabaugh, Pier Luigi Dragotti, Xujiong Ye, Fangde Liu, Simon Arridge, Jennifer Keegan, Yike Guo, David Firmin
2018 IEEE Transactions on Medical Imaging  
Index Terms-Compressed sensing, magnetic resonance imaging (MRI), fast MRI, deep learning, generative adversarial networks (GAN), de-aliasing, inverse problems.  ...  In particular, a novel conditional Generative Adversarial Networks-based model (DAGAN)-based model is proposed to reconstruct CS-MRI.  ...  General GAN Generative Adversarial Networks [52] consist of a generator network G and a discriminator network D.  ... 
doi:10.1109/tmi.2017.2785879 pmid:29870361 fatcat:tqsf5cna2vehjfyu7vppo4i5mu
« Previous Showing results 1 — 15 out of 546 results