80 Hits in 6.0 sec

Validation and Generalizability of Self-Supervised Image Reconstruction Methods for Undersampled MRI [article]

Thomas Yu, Tom Hilbert, Gian Franco Piredda, Arun Joseph, Gabriele Bonanno, Salim Zenkhri, Patrick Omoumi, Meritxell Bach Cuadra, Erick Jorge Canales-Rodríguez, Tobias Kober, Jean-Philippe Thiran
2022 arXiv   pre-print
In this paper, we investigate important aspects of the validation of self-supervised algorithms for reconstruction of undersampled MR images: quantitative evaluation of prospective reconstructions, potential  ...  Deep learning methods have become the state of the art for undersampled MR reconstruction.  ...  National Science Foundation (SNSF, Ambizione grant PZ00P2 185814 to EJC-R) We acknowledge access to the facilities and expertise of the CIBM Center for Biomedical Imaging, a Swiss research center of excellence  ... 
arXiv:2201.12535v3 fatcat:7l36e6ljirer7id6y2pjwwpbj4

Zero-Shot Self-Supervised Learning for MRI Reconstruction [article]

Burhaneddin Yaman, Seyed Amir Hossein Hosseini, Mehmet Akçakaya
2021 arXiv   pre-print
Deep learning (DL) has emerged as a powerful tool for accelerated MRI reconstruction, but these methods often necessitate a database of fully-sampled measurements for training.  ...  In this work, we propose a zero-shot self-supervised learning approach to perform scan-specific accelerated MRI reconstruction to tackle these issues.  ...  Conclusions We proposed a self-supervised zero-shot deep learning method, ZS-SSDU, for scan-specific accelerated MRI reconstruction from a single undersampled dataset.  ... 
arXiv:2102.07737v3 fatcat:amnjsdteabc23nmtpv5glsy3f4

Self-Supervised Learning for MRI Reconstruction with a Parallel Network Training Framework [article]

Chen Hu, Cheng Li, Haifeng Wang, Qiegen Liu, Hairong Zheng, Shanshan Wang
2021 arXiv   pre-print
Image reconstruction from undersampled k-space data plays an important role in accelerating the acquisition of MR data, and a lot of deep learning-based methods have been exploited recently.  ...  To address this issue, we propose a novel self-supervised learning method.  ...  Proposed Self-Supervised Learning Method Fig . 1 shows the overall pipeline of our proposed framework for self-supervised MRI reconstruction.  ... 
arXiv:2109.12502v1 fatcat:flt5pwnj7jaixkh5yemb3ghuki

Self-Score: Self-Supervised Learning on Score-Based Models for MRI Reconstruction [article]

Zhuo-Xu Cui, Chentao Cao, Shaonan Liu, Qingyong Zhu, Jing Cheng, Haifeng Wang, Yanjie Zhu, Dong Liang
2022 arXiv   pre-print
This paper proposes a fully-sampled-data-free score-based diffusion model for MRI reconstruction, which learns the fully sampled MR image prior in a self-supervised manner on undersampled data.  ...  Experiments on the public dataset show that the proposed method outperforms existing self-supervised MRI reconstruction methods and achieves comparable performances with the conventional (fully sampled  ...  This experiment validates the accuracy of the proposed self-supervised learning method for data distribution estimation.  ... 
arXiv:2209.00835v1 fatcat:62y4armqhvaonocilbkdgciwpa

Noise2Recon: A Semi-Supervised Framework for Joint MRI Reconstruction and Denoising [article]

Arjun D Desai, Batu M Ozturkler, Christopher M Sandino, Shreyas Vasanawala, Brian A Hargreaves, Christopher M Re, John M Pauly, Akshay S Chaudhari
2021 arXiv   pre-print
Our method enables the usage of a limited number of fully-sampled and a large number of undersampled-only scans.  ...  Deep learning (DL) has shown promise for faster, high quality accelerated MRI reconstruction.  ...  We then formalize the optimization for supervised MR reconstruction training and for self-supervised denoising training.  ... 
arXiv:2110.00075v1 fatcat:hrk3gcwcifaqhjbznveqazjsna

Self-Supervised Learning of Physics-Guided Reconstruction Neural Networks without Fully-Sampled Reference Data [article]

Burhaneddin Yaman, Seyed Amir Hossein Hosseini, Steen Moeller, Jutta Ellermann, Kâmil Uğurbil, Mehmet Akçakaya
2020 arXiv   pre-print
Theory and Methods: Self-supervised learning via data under-sampling (SSDU) for physics-guided deep learning (DL) reconstruction partitions available measurements into two disjoint sets, one of which is  ...  Purpose: To develop a strategy for training a physics-guided MRI reconstruction neural network without a database of fully-sampled datasets.  ...  Acknowledgements Knee MRI data were obtained from the NYU fastMRI initiative database (58) .  ... 
arXiv:1912.07669v2 fatcat:wlf35xcuszfj7nbzeiyae6afae

VORTEX: Physics-Driven Data Augmentations Using Consistency Training for Robust Accelerated MRI Reconstruction [article]

Arjun D Desai, Beliz Gunel, Batu M Ozturkler, Harris Beg, Shreyas Vasanawala, Brian A Hargreaves, Christopher Ré, John M Pauly, Akshay S Chaudhari
2022 arXiv   pre-print
regimes; (2) considerably outperforms state-of-the-art purely image-based data augmentation techniques and self-supervised reconstruction methods on both in-distribution and out-of-distribution data;  ...  Deep neural networks have enabled improved image quality and fast inference times for various inverse problems, including accelerated magnetic resonance imaging (MRI) reconstruction.  ...  in Medicine and Imaging GCP grant; Stanford Human-Centered Artificial Intelligence GCP grant; GE Healthcare and Philips.  ... 
arXiv:2111.02549v2 fatcat:wy2wxgo6ive6hnat6we4yn4wtm

Unsupervised MRI Reconstruction via Zero-Shot Learned Adversarial Transformers [article]

Yilmaz Korkmaz, Salman UH Dar, Mahmut Yurt, Muzaffer Özbey, Tolga Çukur
2022 arXiv   pre-print
Supervised reconstruction models are characteristically trained on matched pairs of undersampled and fully-sampled data to capture an MRI prior, along with supervision regarding the imaging operator to  ...  During inference, a zero-shot reconstruction is then performed by incorporating the imaging operator and optimizing the prior to maximize consistency to undersampled data.  ...  Benefits of SLATER over state-of-the-art supervised and unsupervised methods were demonstrated in brain MRI.  ... 
arXiv:2105.08059v3 fatcat:wdkx56xacfay3ib5mnxv4viwuu

Real‐time 3D motion estimation from undersampled MRI using multi‐resolution neural networks

M. L. Terpstra, M. Maspero, T. Bruijnen, J.J.C. Verhoeff, J.J.W. Lagendijk, C.A.T. van den Berg
2021 Medical Physics (Lancaster)  
Theory and Methods: Respiratory-resolved T 1 -weighted 4D-MRI of 27 patients with lung cancer were acquired using a golden-angle radial stack-of-stars readout.  ...  TEMPEST was validated using 4D respiratory-resolved MRI, a digital phantom, and a physical motion phantom.  ...  METHODS We trained a supervised multiresolution DL model (TEMPEST) to estimate a DVF (DVF TEMPEST ) between two undersampled MRI volumes acquired with a goldenangle radial stack-of -stars readout.  ... 
doi:10.1002/mp.15217 pmid:34525223 pmcid:PMC9298075 fatcat:b7m25l2rkndediybmynbrssfwe

Multi-head Cascaded Swin Transformers with Attention to k-space Sampling Pattern for Accelerated MRI Reconstruction [article]

Mevan Ekanayake, Kamlesh Pawar, Mehrtash Harandi, Gary Egan, Zhaolin Chen
2022 arXiv   pre-print
The self-attention-based transformer models are capable of capturing global correlations among image features, however, the current contributions of transformer models for MRI reconstruction are minute  ...  Our model significantly outperforms state-of-the-art MRI reconstruction methods both visually and quantitatively while depicting improved resolution and removal of aliasing artifacts.  ...  Acknowledgments The authors are grateful for support from Australian Research Council Linkage grant LP170100494 and Australian Research Council Discovery grant DP210101863.  ... 
arXiv:2207.08412v1 fatcat:ewu3rsg2tvck7oybmcua4smjje

Multi-Coil MRI Reconstruction Challenge – Assessing Brain MRI Reconstruction Models and their Generalizability to Varying Coil Configurations [article]

Youssef Beauferris, Jonas Teuwen, Dimitrios Karkalousos, Nikita Moriakov, Mattha Caan, George Yiasemis, Lívia Rodrigues, Alexandre Lopes, Hélio Pedrini, Letícia Rittner, Maik Dannecker, Viktor Studenyak (+11 others)
2021 arXiv   pre-print
Deep-learning-based brain magnetic resonance imaging (MRI) reconstruction methods have the potential to accelerate the MRI acquisition process.  ...  They provide an objective performance assessment for future developments in the field of brain MRI reconstruction.  ...  The organizers of the challenge also acknowledge Nvidia for providing a Titan V graphics processing unit and Amazon Web Services for providing computational infra-structure that was used by some of the  ... 
arXiv:2011.07952v2 fatcat:t55wnasrpveuzio2kfne756fhq

Accelerated coronary MRI with sRAKI: A database-free self-consistent neural network k-space reconstruction for arbitrary undersampling

Seyed Amir Hossein Hosseini, Chi Zhang, Sebastian Weingärtner, Steen Moeller, Matthias Stuber, Kamil Ugurbil, Mehmet Akçakaya, Ulas Bagci
2020 PLoS ONE  
The data were retrospectively undersampled, and reconstructed using SPIRiT, l1-SPIRiT and sRAKI for acceleration rates of 2 to 5.  ...  Self-consistent robust artificial-neural-networks for k-space interpolation (sRAKI) performs iterative parallel imaging reconstruction by enforcing self-consistency among coils.  ...  Supervision: Kamil Ugurbil, Mehmet Akçakaya. Validation: Mehmet Akçakaya.  ... 
doi:10.1371/journal.pone.0229418 pmid:32084235 fatcat:t6tzg3vcnzgt3pyhx3245tnrfq

Machine Learning in Magnetic Resonance Imaging: Image Reconstruction [article]

Javier Montalt-Tordera, Vivek Muthurangu, Andreas Hauptmann, Jennifer Anne Steeden
2020 arXiv   pre-print
Promising results have been demonstrated from a range of methods, enabling natural looking images and rapid computation.  ...  Following the success of machine learning in a wide range of imaging tasks, there has been a recent explosion in the use of machine learning in the field of MRI image reconstruction.  ...  There have been many approaches to machine learning MRI reconstruction (both in terms of supervised and unsupervised techniques), including methods which work in image-space, those which work in k-space  ... 
arXiv:2012.05303v1 fatcat:uqkxn3cqxff4nm3t5nbdia44iq

Adaptive Local Neighborhood-based Neural Networks for MR Image Reconstruction from Undersampled Data [article]

Shijun Liang, Anish Lahiri, Saiprasad Ravishankar
2022 arXiv   pre-print
Reconstructions were performed at four fold and eight fold undersampling of k-space with 1D variable-density random phase-encode undersampling masks.  ...  Our approach, dubbed LONDN-MRI, was validated on the FastMRI multi-coil knee data set using deep unrolled reconstruction networks.  ...  A related approach dubbed self-supervised learning has also shown promise for MRI [35] and uses a large unpaired data set. A.  ... 
arXiv:2206.00775v1 fatcat:klj7c7766zf3toqigybakz35y4

A Transfer-Learning Approach for Accelerated MRI using Deep Neural Networks [article]

Salman Ul Hassan Dar, Muzaffer Özbey, Ahmet Burak Çatlı, Tolga Çukur
2019 arXiv   pre-print
Purpose: Neural networks have received recent interest for reconstruction of undersampled MR acquisitions.  ...  Conclusion: The proposed approach might facilitate the use of neural networks for MRI reconstruction without the need for collection of extensive imaging datasets.  ...  Another method to enhance generalizability was to compound datasets containing a mixture of distinct MRI contrasts during network training (29) .  ... 
arXiv:1710.02615v3 fatcat:m3i65su4hnbcnfdvzysdahf264
« Previous Showing results 1 — 15 out of 80 results