Filters








46 Hits in 2.3 sec

fastMRI: An Open Dataset and Benchmarks for Accelerated MRI [article]

Jure Zbontar, Florian Knoll, Anuroop Sriram, Tullie Murrell, Zhengnan Huang, Matthew J. Muckley, Aaron Defazio, Ruben Stern, Patricia Johnson, Mary Bruno, Marc Parente, Krzysztof J. Geras (+14 others)
2019 arXiv   pre-print
We introduce the fastMRI dataset, a large-scale collection of both raw MR measurements and clinical MR images, that can be used for training and evaluation of machine-learning approaches to MR image reconstruction  ...  Accelerating Magnetic Resonance Imaging (MRI) by taking fewer measurements has the potential to reduce medical costs, minimize stress to patients and make MRI possible in applications where it is currently  ...  The fastMRI Dataset and Associated Tasks The fastMRI dataset (http://fastmri.med.nyu.edu/) contains four types of data from MRI acquisitions of knees and brains Raw multi-coil k-space data: unprocessed  ... 
arXiv:1811.08839v2 fatcat:wyfbjvimbzcy3llffqagrea24e

DIRECT: Deep Image REConstruction Toolkit

George Yiasemis, Nikita Moriakov, Dimitrios Karkalousos, Matthan Caan, Jonas Teuwen
2022 Journal of Open Source Software  
using DIRECT include Yiasemis et al. (2022) (presented in SPIE Medical Imaging Conference 2022) and Yiasemis et al. (2021) (to be presented in CVPR Conference 2022). fastMRI: An open dataset and benchmarks  ...  Challenges DIRECT has been used for MRI Reconstruction result submissions in the fastMRI challenge (Muckley et al., 2021) and the Multi-Coil MRI Reconstruction challenge (Beauferris et al., 2021) .  ... 
doi:10.21105/joss.04278 fatcat:eynbtq4adjdoncm2skwhbk3ggu

Towards Ultrafast MRI via Extreme k-Space Undersampling and Superresolution [article]

Aleksandr Belov and Joel Stadelmann and Sergey Kastryulin and Dmitry V. Dylov
2021 arXiv   pre-print
We went below the MRI acceleration factors (a.k.a., k-space undersampling) reported by all published papers that reference the original fastMRI challenge, and then considered powerful deep learning based  ...  The quality of the reconstructed images surpasses that of the other methods, yielding an MSE of 0.00114, a PSNR of 29.6 dB, and an SSIM of 0.956 at x16 acceleration factor.  ...  At ×16 acceleration, SRGAN+U-Net achieved an MSE of 11.4 · 10 −4 , a PSNR of 29.6 dB, and an SSIM of 0.956 on the fastMRI brain data.  ... 
arXiv:2103.02940v1 fatcat:gpiwa77q3ve3bhuagnkudnqoye

Benchmarking MRI Reconstruction Neural Networks on Large Public Datasets

Zaccharie Ramzi, Philippe Ciuciu, Jean-Luc Starck
2020 Applied Sciences  
This paper shows the results obtained for this benchmark, allowing to compare the networks, and links the open source implementation of all these networks in Keras.  ...  The recent release of a public dataset, fastMRI, consisting of raw k-space data, encouraged us to write a consistent benchmark of several deep neural networks for MR image reconstruction.  ...  Acknowledgments: We want to thank Jonas Adler, Justin Haldar, and Jo Schlemper for their very useful and benevolent remarks and answers they gave when asked questions about their works.  ... 
doi:10.3390/app10051816 fatcat:3pbq3c6x4ffxdjaphzsqexezam

Optimal MRI Undersampling Patterns for Ultimate Benefit of Medical Vision Tasks [article]

Artem Razumov, Oleg Y. Rogov, Dmitry V. Dylov
2021 arXiv   pre-print
We validate the proposed MRI acceleration paradigm on three classical medical datasets, demonstrating a noticeable improvement of the target metrics at the high acceleration factors (for the segmentation  ...  To accelerate MRI, the field of compressed sensing is traditionally concerned with optimizing the image quality after a partial undersampling of the measurable k-space.  ...  In-depth reviews [47] and [48] describe validation of MRI acceleration methods on BraTS and FastMRI datasets; and the review [49] covers state-of-the-art (SOTA) for the ACDC data. B.  ... 
arXiv:2108.04914v1 fatcat:muqf3kexzjetbo7og6yminj3ei

HUMUS-Net: Hybrid unrolled multi-scale network architecture for accelerated MRI reconstruction [article]

Zalan Fabian, Mahdi Soltanolkotabi
2022 arXiv   pre-print
Our network establishes new state of the art on the largest publicly available MRI dataset, the fastMRI dataset.  ...  In accelerated MRI reconstruction, the anatomy of a patient is recovered from a set of under-sampled and noisy measurements.  ...  More recently, the fastMRI dataset [Zbontar et al., 2019] , the largest publicly available MRI dataset, has been gaining ground as a standard benchmark to evaluate MRI reconstruction methods.  ... 
arXiv:2203.08213v1 fatcat:vhjiteq3xfdhjoal3xj2j37im4

Data augmentation for deep learning based accelerated MRI reconstruction with limited data [article]

Zalan Fabian, Reinhard Heckel, Mahdi Soltanolkotabi
2021 arXiv   pre-print
Inspired by the success of Data Augmentation (DA) for classification problems, in this paper, we propose a pipeline for data augmentation for accelerated MRI reconstruction and study its effectiveness  ...  Deep neural networks have emerged as very successful tools for image restoration and reconstruction tasks.  ...  Soltanolkotabi is supported by the Packard Fellowship in Science and Engineering, a Sloan Research Fellowship in Mathematics, an NSF-CAREER under award #1846369, DARPA Learning with Less Labels (LwLL)  ... 
arXiv:2106.14947v1 fatcat:7rflgunr3jfdzazdb4txfalaea

Systematic evaluation of iterative deep neural networks for fast parallel MRI reconstruction with sensitivity-weighted coil combination

Kerstin Hammernik, Jo Schlemper, Chen Qin, Jinming Duan, Ronald M Summers, Daniel Rueckert
2021 Magnetic Resonance in Medicine  
In this work, we study how dataset sizes affect single-anatomy and cross-anatomy training of neural networks for MRI reconstruction.  ...  The proposed networks are compared to other state-of-the-art approaches on the publicly available fastMRI knee and neuro dataset and tested for stability across different training configurations regarding  ...  All networks trained on knee data fail for this dataset and acceleration factor.  ... 
doi:10.1002/mrm.28827 pmid:34110037 fatcat:5hxg3kmzzrabfokszzric62lzm

Learning-Based Optimization of the Under-Sampling Pattern in MRI [chapter]

Cagla Deniz Bahadir, Adrian V. Dalca, Mert R. Sabuncu
2019 Lecture Notes in Computer Science  
For a provided sparsity constraint, our method optimizes the sub-sampling pattern and reconstruction model, using an end-to-end unsupervised learning strategy.  ...  The long scan times of Magnetic Resonance Imaging (MRI) create a bottleneck in patient care and acquisitions can be accelerated by under-sampling in k-space (i.e., the Fourier domain).  ...  NYU fastMRI We also conducted experiments on NYU fastMRI dataset [35] ; which is an openly available, large-scale, public data set of raw k-space data of knee scans.  ... 
doi:10.1007/978-3-030-20351-1_61 fatcat:vy6ojeq3d5bwjnct4s3bertk3a

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).  ...  Moreover, both trained and un-trained methods tuned for a particular dataset suffer very similarly from distribution shifts.  ...  “fastMRI: An open dataset and benchmarks for accelerated MRI”.  ... 
arXiv:2102.06103v2 fatcat:57p7hrq2abeqbb4x2ufntj2274

Density Compensated Unrolled Networks for Non-Cartesian MRI Reconstruction [article]

Zaccharie Ramzi, Jean-Luc Starck, Philippe Ciuciu
2021 arXiv   pre-print
We assess their efficiency on the publicly available fastMRI dataset, and perform a small ablation study.  ...  We also open source our code, in particular a Non-Uniform Fast Fourier transform for TensorFlow.  ...  ACKNOWLEDGEMENTS We are grateful to Jo Schlemper for his very useful remarks and answers to our questions. We also thank Chaithya G.R. for the discussions about DC.  ... 
arXiv:2101.01570v2 fatcat:cdhk3zchefflxop7nvu5kqjzum

An Adaptive Intelligence Algorithm for Undersampled Knee MRI Reconstruction

Nicola Pezzotti, Sahar Yousefi, Mohamed S. Elmahdy, Jeroen van Gemert, Christophe Schulke, Mariya Doneva, Tim Nielsen, Sergey Kastryulin, Boudewijn P.F. Lelieveldt, Boudewijn P.F. Lelieveldt, Matthias J.P. van Osch, Elwin de Weerdt (+1 others)
2020 IEEE Access  
The network was trained and tested on a knee MRI dataset from the 2019 fastMRI challenge organized by Facebook AI Research and NYU Langone Health.  ...  Our method was evaluated by the fastMRI organizers on an independent challenge dataset.  ...  [24] provided a reproducible benchmark of deep learning based reconstruction methods on the single-coil part of the fastMRI dataset [25] .  ... 
doi:10.1109/access.2020.3034287 fatcat:rrzcxgsxkjbgne2sjditqp5biy

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
They provide an objective performance assessment for future developments in the field of brain MRI reconstruction.  ...  The MC-MRI benchmark data, evaluation code and current challenge leaderboard are publicly available.  ...  Acknowledgements Richard Frayne thanks the Canadian Institutes for Health Research (CIHR, FDN-143298) for supporting the Calgary Normative Study and acquiring the raw datasets.  ... 
arXiv:2011.07952v2 fatcat:t55wnasrpveuzio2kfne756fhq

SKM-TEA: A Dataset for Accelerated MRI Reconstruction with Dense Image Labels for Quantitative Clinical Evaluation [article]

Arjun D Desai, Andrew M Schmidt, Elka B Rubin, Christopher M Sandino, Marianne S Black, Valentina Mazzoli, Kathryn J Stevens, Robert Boutin, Christopher Ré, Garry E Gold, Brian A Hargreaves, Akshay S Chaudhari
2022 arXiv   pre-print
Dataset access, code, and benchmarks are available at https://github.com/StanfordMIMI/skm-tea.  ...  Finally, we use this framework to benchmark state-of-the-art baselines on this dataset.  ...  Related Work Previous datasets for MRI reconstruction, segmentation, and classification have been essential for enabling ML benchmarks for MRI research.  ... 
arXiv:2203.06823v1 fatcat:twiuafgth5arpefpg7z4mhktku

Brain MRI reconstruction challenge with realistic noise [article]

Melanie Ganz, Hannah Eichhorn
2021 Zenodo  
While the fastMRI challenge provides participants with a large dataset, the way the under-sampled k-space data is constructed is artificial and not realistic, since patient motion corrupts the k-space  ...  We propose a challenge that aims at checking the robustness of fast MRI reconstruction methods by providing image and k-space datasets that consist of ground truth as well as motion degraded scans.  ...  We will compare the SSIM values used for ranking with values for comparable image metrics like Peak Signal-to-Noise ration or Tenengrad, as well as the visual assessments by the radiologists.  ... 
doi:10.5281/zenodo.4572639 fatcat:b243wms6tvhqjn43ulxblkteam
« Previous Showing results 1 — 15 out of 46 results