Results of the 2020 fastMRI Challenge for Machine Learning MR Image Reconstruction (2021-04-30T00:00:00.000000Z)

TL;DR

This paper hosts the second fastMRI competition targeted towards reconstructing MR images with subsampled k-space data and identifies common failure modes across the submissions, highlighting areas of need for future research in the MRI reconstruction community.

Abstract

Accelerating MRI scans is one of the principal outstanding problems in the MRI research community. Towards this goal, we hosted the second fastMRI competition targeted towards reconstructing MR images with subsampled k-space data. We provided participants with data from 7,299 clinical brain scans (de-identified via a HIPAA-compliant procedure by NYU Langone Health), holding back the fully-sampled data from 894 of these scans for challenge evaluation purposes. In contrast to the 2019 challenge, we focused our radiologist evaluations on pathological assessment in brain images. We also debuted a new Transfer track that required participants to submit models evaluated on MRI scanners from outside the training set. We received 19 submissions from eight different groups. Results showed one team scoring best in both SSIM scores and qualitative radiologist evaluations. We also performed analysis on alternative metrics to mitigate the effects of background noise and collected feedback from the participants to inform future challenges. Lastly, we identify common failure modes across the submissions, highlighting areas of need for future research in the MRI reconstruction community.

TL;DR

Abstract

Authors

References66 items

Joint Deep Model-based MR Image and Coil Sensitivity Reconstruction Network (Joint-ICNet) for Fast MRI

Multi-channel MR Reconstruction (MC-MRRec) Challenge - Comparing Accelerated MR Reconstruction Models and Assessing Their Genereralizability to Datasets Collected with Different Coils

XPDNet for MRI Reconstruction: an application to the 2020 fastMRI challenge

CG‐SENSE revisited: Results from the first ISMRM reproducibility challenge

Using Deep Learning to Accelerate Knee MRI at 3T: Results of an Interchangeability Study.

An Adaptive Intelligence Algorithm for Undersampled Knee MRI Reconstruction

End-to-End Variational Networks for Accelerated MRI Reconstruction

Benchmarking MRI Reconstruction Neural Networks on Large Public Datasets

fastMRI: A Publicly Available Raw k-Space and DICOM Dataset of Knee Images for Accelerated MR Image Reconstruction Using Machine Learning.

MRI Banding Removal via Adversarial Training

Advancing machine learning for MR image reconstruction with an open competition: Overview of the 2019 fastMRI challenge

Σ-net: Systematic Evaluation of Iterative Deep Neural Networks for Fast Parallel MR Image Reconstruction

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

PyTorch: An Imperative Style, High-Performance Deep Learning Library

Pyramid Convolutional RNN for MRI Reconstruction

i-RIM applied to the fastMRI challenge

On the Variance of the Adaptive Learning Rate and Beyond

Deep Iterative Down-Up CNN for Image Denoising

Deep-Learning Methods for Parallel Magnetic Resonance Imaging Reconstruction: A Survey of the Current Approaches, Trends, and Issues

Challenges in diffusion MRI tractography - Lessons learned from international benchmark competitions.

On instabilities of deep learning in image reconstruction and the potential costs of AI

Recurrent inference machines for reconstructing heterogeneous MRI data

Assessment of the generalization of learned image reconstruction and the potential for transfer learning

fastMRI: An Open Dataset and Benchmarks for Accelerated MRI

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

Multi-level Wavelet-CNN for Image Restoration

KIKI‐net: cross‐domain convolutional neural networks for reconstructing undersampled magnetic resonance images

MoDL: Model-Based Deep Learning Architecture for Inverse Problems

Advancing RF pulse design using an open‐competition format: Report from the 2015 ISMRM challenge

Learned Primal-Dual Reconstruction

Recurrent Inference Machines for Solving Inverse Problems

Image reconstruction by domain-transform manifold learning

A Deep Cascade of Convolutional Neural Networks for Dynamic MR Image Reconstruction

Learning a variational network for reconstruction of accelerated MRI data

Deep ADMM-Net for Compressive Sensing MRI

Densely Connected Convolutional Networks

Deep Residual Learning for Image Recognition

Rethinking the Inception Architecture for Computer Vision

U-Net: Convolutional Networks for Biomedical Image Segmentation

Adam: A Method for Stochastic Optimization

cuDNN: Efficient Primitives for Deep Learning

Very Deep Convolutional Networks for Large-Scale Image Recognition

ESPIRiT—an eigenvalue approach to autocalibrating parallel MRI: Where SENSE meets GRAPPA

ImageNet classification with deep convolutional neural networks

A First-Order Primal-Dual Algorithm for Convex Problems with Applications to Imaging

Deep, Big, Simple Neural Nets for Handwritten Digit Recognition

ImageNet: A large-scale hierarchical image database

Large-scale deep unsupervised learning using graphics processors

Compressed Sensing MRI

Image quality assessment: from error visibility to structural similarity

Stochastic resonance and sensory information processing: a tutorial and review of application

Generalized autocalibrating partially parallel acquisitions (GRAPPA)

Adaptive reconstruction of phased array MR imagery

SENSE: Sensitivity encoding for fast MRI

Simultaneous acquisition of spatial harmonics (SMASH): Fast imaging with radiofrequency coil arrays

The NMR phased array

Neocognitron: A self-organizing neural network model for a mechanism of pattern recognition unaffected by shift in position

Certain Topics in Telegraph Transmission Theory

Berkeley advanced reconstruction toolbox,

Gradient-based learning applied to document recognition

Neocognitron: A Self-Organizing Neural Network Model for a Mechanism of Visual Pattern Recognition

Communication in the Presence of Noise

“On the carrying capacity of the ‘ether’ and wire in telecommunications

XVIII.—On the Functions which are represented by the Expansions of the Interpolation-Theory

Distributed under Creative Commons Cc-by 4.0 Scikit-image: Image Processing in Python

This Paper Is Included in the Proceedings of the 12th Usenix Symposium on Operating Systems Design and Implementation (osdi '16). Tensorflow: a System for Large-scale Machine Learning Tensorflow: a System for Large-scale Machine Learning